HW4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f99ac30c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fa6c9fe2",
    "outputId": "ef75c361-0179-4fd1-d052-4da981cf6bcb"
   },
   "source": [
    "# Colab setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "81abf1af",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from google.colab import drive\n",
    "# drive.mount(\"/content/gdrive\", force_remount=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "y5jAfMMw9zNc",
   "metadata": {
    "id": "y5jAfMMw9zNc"
   },
   "outputs": [],
   "source": [
    "# from google.colab import files\n",
    "# files.upload()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "722Z128J99Rb",
   "metadata": {
    "id": "722Z128J99Rb"
   },
   "outputs": [],
   "source": [
    "# !ls -lha kaggle.json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "KZcpr5Zg-CfV",
   "metadata": {
    "id": "KZcpr5Zg-CfV"
   },
   "outputs": [],
   "source": [
    "# !pip install -q kaggle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "-nN8_Lkz-Fr6",
   "metadata": {
    "id": "-nN8_Lkz-Fr6"
   },
   "outputs": [],
   "source": [
    "# !mkdir -p ~/.kaggle\n",
    "# !cp kaggle.json ~/.kaggle/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "G1W3RdUX-IID",
   "metadata": {
    "id": "G1W3RdUX-IID"
   },
   "outputs": [],
   "source": [
    "# !chmod 600 /root/.kaggle/kaggle.json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "Bd9mPwxS-KZw",
   "metadata": {
    "id": "Bd9mPwxS-KZw"
   },
   "outputs": [],
   "source": [
    "# !pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "g17VuxOx_PVu",
   "metadata": {
    "id": "g17VuxOx_PVu"
   },
   "outputs": [],
   "source": [
    "# !kaggle datasets download -d kazanova/sentiment140"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "sCmoRMd0_ZhX",
   "metadata": {
    "id": "sCmoRMd0_ZhX"
   },
   "outputs": [],
   "source": [
    "# !unzip sentiment140.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "-P0hOpL2AvVU",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-P0hOpL2AvVU",
    "outputId": "8ee9e5b9-37cf-4668-ecd8-46ba476ed7e6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gdrive\tsample_data\n"
     ]
    }
   ],
   "source": [
    "!ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "NKk1EXCVAyhX",
   "metadata": {
    "id": "NKk1EXCVAyhX"
   },
   "outputs": [],
   "source": [
    "# !mv training.1600000.processed.noemoticon.csv data.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "Q_4THVsKA4jX",
   "metadata": {
    "id": "Q_4THVsKA4jX"
   },
   "outputs": [],
   "source": [
    "# !ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "uYUwDDedWkh2",
   "metadata": {
    "id": "uYUwDDedWkh2"
   },
   "outputs": [],
   "source": [
    "!cp /content/gdrive/MyDrive/MIR-HW4/sentiment140.zip /content/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "98HM8GemXDvJ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "98HM8GemXDvJ",
    "outputId": "7d0560e8-75af-446b-942f-83e265d4da64"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Archive:  sentiment140.zip\n",
      "  inflating: training.1600000.processed.noemoticon.csv  \n"
     ]
    }
   ],
   "source": [
    "!unzip sentiment140.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "_UYRH3r-XLDr",
   "metadata": {
    "id": "_UYRH3r-XLDr"
   },
   "outputs": [],
   "source": [
    "!mv training.1600000.processed.noemoticon.csv data.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "ICWyk0J5A_bu",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ICWyk0J5A_bu",
    "outputId": "ce2bfbd0-4e06-4568-e80f-925daa42254d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
      "Collecting unidecode\n",
      "  Downloading Unidecode-1.3.4-py3-none-any.whl (235 kB)\n",
      "\u001b[K     |████████████████████████████████| 235 kB 29.2 MB/s \n",
      "\u001b[?25hInstalling collected packages: unidecode\n",
      "Successfully installed unidecode-1.3.4\n",
      "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
      "Collecting autocorrect\n",
      "  Downloading autocorrect-2.6.1.tar.gz (622 kB)\n",
      "\u001b[K     |████████████████████████████████| 622 kB 37.3 MB/s \n",
      "\u001b[?25hBuilding wheels for collected packages: autocorrect\n",
      "  Building wheel for autocorrect (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
      "  Created wheel for autocorrect: filename=autocorrect-2.6.1-py3-none-any.whl size=622382 sha256=48ed1372f932bba0cd306ead9ecebad347577e0e1dc0c8e8cc0434c9af7618b2\n",
      "  Stored in directory: /root/.cache/pip/wheels/54/d4/37/8244101ad50b0f7d9bffd93ce58ed7991ee1753b290923934b\n",
      "Successfully built autocorrect\n",
      "Installing collected packages: autocorrect\n",
      "Successfully installed autocorrect-2.6.1\n",
      "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
      "Collecting transformers\n",
      "  Downloading transformers-4.20.1-py3-none-any.whl (4.4 MB)\n",
      "\u001b[K     |████████████████████████████████| 4.4 MB 30.0 MB/s \n",
      "\u001b[?25hRequirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2022.6.2)\n",
      "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.64.0)\n",
      "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.23.0)\n",
      "Requirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.7.1)\n",
      "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.21.6)\n",
      "Collecting tokenizers!=0.11.3,<0.13,>=0.11.1\n",
      "  Downloading tokenizers-0.12.1-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (6.6 MB)\n",
      "\u001b[K     |████████████████████████████████| 6.6 MB 61.6 MB/s \n",
      "\u001b[?25hCollecting huggingface-hub<1.0,>=0.1.0\n",
      "  Downloading huggingface_hub-0.8.1-py3-none-any.whl (101 kB)\n",
      "\u001b[K     |████████████████████████████████| 101 kB 12.0 MB/s \n",
      "\u001b[?25hRequirement already satisfied: importlib-metadata in /usr/local/lib/python3.7/dist-packages (from transformers) (4.11.4)\n",
      "Collecting pyyaml>=5.1\n",
      "  Downloading PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (596 kB)\n",
      "\u001b[K     |████████████████████████████████| 596 kB 74.5 MB/s \n",
      "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.7/dist-packages (from transformers) (21.3)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from huggingface-hub<1.0,>=0.1.0->transformers) (4.1.1)\n",
      "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=20.0->transformers) (3.0.9)\n",
      "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata->transformers) (3.8.0)\n",
      "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3)\n",
      "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10)\n",
      "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2022.6.15)\n",
      "Installing collected packages: pyyaml, tokenizers, huggingface-hub, transformers\n",
      "  Attempting uninstall: pyyaml\n",
      "    Found existing installation: PyYAML 3.13\n",
      "    Uninstalling PyYAML-3.13:\n",
      "      Successfully uninstalled PyYAML-3.13\n",
      "Successfully installed huggingface-hub-0.8.1 pyyaml-6.0 tokenizers-0.12.1 transformers-4.20.1\n"
     ]
    }
   ],
   "source": [
    "!pip install unidecode\n",
    "!pip install autocorrect\n",
    "!pip install transformers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "26b20c1a",
   "metadata": {},
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "RI22tWnD9su_",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RI22tWnD9su_",
    "outputId": "f5a273de-41a3-434d-9d67-25adf7fbca5f"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
      "[nltk_data]   Unzipping tokenizers/punkt.zip.\n",
      "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
      "[nltk_data]   Unzipping corpora/wordnet.zip.\n",
      "[nltk_data] Downloading package omw-1.4 to /root/nltk_data...\n",
      "[nltk_data]   Unzipping corpora/omw-1.4.zip.\n",
      "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
      "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "device(type='cuda', index=0)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "import unidecode \n",
    "import pandas as pd \n",
    "import re \n",
    "import time \n",
    "import nltk \n",
    "nltk.download('punkt')\n",
    "nltk.download('wordnet')\n",
    "nltk.download('omw-1.4')\n",
    "\n",
    "from nltk.corpus import stopwords\n",
    "nltk.download('stopwords')\n",
    "from nltk.tokenize import word_tokenize \n",
    "from nltk.stem import WordNetLemmatizer\n",
    "from autocorrect import Speller\n",
    "from bs4 import BeautifulSoup\n",
    "from nltk.corpus import stopwords \n",
    "from nltk import word_tokenize\n",
    "import string\n",
    "import json\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "\n",
    "import transformers\n",
    "from transformers import BertModel, BertConfig, BertTokenizer, AdamW, get_linear_schedule_with_warmup\n",
    "import torch\n",
    "\n",
    "from pylab import rcParams\n",
    "from matplotlib import rc\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from collections import defaultdict\n",
    "from textwrap import wrap\n",
    "\n",
    "from torch import nn, optim\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "import torch.nn.functional as F\n",
    "\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format='retina'\n",
    "\n",
    "sns.set(style='whitegrid', palette='muted', font_scale=1.2)\n",
    "HAPPY_COLORS_PALETTE = [\"#01BEFE\", \"#FFDD00\", \"#FF7D00\", \"#FF006D\", \"#ADFF02\", \"#8F00FF\"]\n",
    "sns.set_palette(sns.color_palette(HAPPY_COLORS_PALETTE))\n",
    "rcParams['figure.figsize'] = 12, 8\n",
    "\n",
    "from tqdm import tqdm\n",
    "\n",
    "RANDOM_SEED = 76\n",
    "np.random.seed(RANDOM_SEED)\n",
    "torch.manual_seed(RANDOM_SEED)\n",
    "\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "device"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27fadfd0",
   "metadata": {},
   "source": [
    "# Reading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "5584a282",
   "metadata": {
    "id": "5584a282"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"data.csv\", header=None, encoding=\"latin-1\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c5f8cd69",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "c5f8cd69",
    "outputId": "5a22d4e4-82c5-47e3-d1d1-835f6ba579f7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1600000, 6)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d7b3d056",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "d7b3d056",
    "outputId": "d6a01202-2c49-4940-ad9a-75743a783a30"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-ff331995-9c36-40c1-80d2-3bb4e5c23ce5\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810369</td>\n",
       "      <td>Mon Apr 06 22:19:45 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>_TheSpecialOne_</td>\n",
       "      <td>@switchfoot http://twitpic.com/2y1zl - Awww, t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810672</td>\n",
       "      <td>Mon Apr 06 22:19:49 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>scotthamilton</td>\n",
       "      <td>is upset that he can't update his Facebook by ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810917</td>\n",
       "      <td>Mon Apr 06 22:19:53 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>mattycus</td>\n",
       "      <td>@Kenichan I dived many times for the ball. Man...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811184</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>ElleCTF</td>\n",
       "      <td>my whole body feels itchy and like its on fire</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811193</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>Karoli</td>\n",
       "      <td>@nationwideclass no, it's not behaving at all....</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ff331995-9c36-40c1-80d2-3bb4e5c23ce5')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-ff331995-9c36-40c1-80d2-3bb4e5c23ce5 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-ff331995-9c36-40c1-80d2-3bb4e5c23ce5');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   0           1                             2         3                4  \\\n",
       "0  0  1467810369  Mon Apr 06 22:19:45 PDT 2009  NO_QUERY  _TheSpecialOne_   \n",
       "1  0  1467810672  Mon Apr 06 22:19:49 PDT 2009  NO_QUERY    scotthamilton   \n",
       "2  0  1467810917  Mon Apr 06 22:19:53 PDT 2009  NO_QUERY         mattycus   \n",
       "3  0  1467811184  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY          ElleCTF   \n",
       "4  0  1467811193  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY           Karoli   \n",
       "\n",
       "                                                   5  \n",
       "0  @switchfoot http://twitpic.com/2y1zl - Awww, t...  \n",
       "1  is upset that he can't update his Facebook by ...  \n",
       "2  @Kenichan I dived many times for the ball. Man...  \n",
       "3    my whole body feels itchy and like its on fire   \n",
       "4  @nationwideclass no, it's not behaving at all....  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "fec40f9b",
   "metadata": {
    "id": "fec40f9b"
   },
   "outputs": [],
   "source": [
    "df.columns = ['target', 'tweet_id', 'datetime', 'query', 'id', 'tweet']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "703e3f7a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "703e3f7a",
    "outputId": "4a466179-752a-41ea-f77d-7cfd1949f57e"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-2a7fb3e9-b024-41af-a318-8d458e5d8dd2\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>target</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>datetime</th>\n",
       "      <th>query</th>\n",
       "      <th>id</th>\n",
       "      <th>tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810369</td>\n",
       "      <td>Mon Apr 06 22:19:45 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>_TheSpecialOne_</td>\n",
       "      <td>@switchfoot http://twitpic.com/2y1zl - Awww, t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810672</td>\n",
       "      <td>Mon Apr 06 22:19:49 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>scotthamilton</td>\n",
       "      <td>is upset that he can't update his Facebook by ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810917</td>\n",
       "      <td>Mon Apr 06 22:19:53 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>mattycus</td>\n",
       "      <td>@Kenichan I dived many times for the ball. Man...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811184</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>ElleCTF</td>\n",
       "      <td>my whole body feels itchy and like its on fire</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811193</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>Karoli</td>\n",
       "      <td>@nationwideclass no, it's not behaving at all....</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a7fb3e9-b024-41af-a318-8d458e5d8dd2')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-2a7fb3e9-b024-41af-a318-8d458e5d8dd2 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-2a7fb3e9-b024-41af-a318-8d458e5d8dd2');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   target    tweet_id                      datetime     query  \\\n",
       "0       0  1467810369  Mon Apr 06 22:19:45 PDT 2009  NO_QUERY   \n",
       "1       0  1467810672  Mon Apr 06 22:19:49 PDT 2009  NO_QUERY   \n",
       "2       0  1467810917  Mon Apr 06 22:19:53 PDT 2009  NO_QUERY   \n",
       "3       0  1467811184  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY   \n",
       "4       0  1467811193  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY   \n",
       "\n",
       "                id                                              tweet  \n",
       "0  _TheSpecialOne_  @switchfoot http://twitpic.com/2y1zl - Awww, t...  \n",
       "1    scotthamilton  is upset that he can't update his Facebook by ...  \n",
       "2         mattycus  @Kenichan I dived many times for the ball. Man...  \n",
       "3          ElleCTF    my whole body feels itchy and like its on fire   \n",
       "4           Karoli  @nationwideclass no, it's not behaving at all....  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "821634e6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "821634e6",
    "outputId": "340f6c3c-2255-4c50-b586-a12f949b084f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "659775"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df.id.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "843159a6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "843159a6",
    "outputId": "1ee07a00-ebce-42ea-c441-7c3caca5a47f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "@switchfoot http://twitpic.com/2y1zl - Awww, that's a bummer.  You shoulda got David Carr of Third Day to do it. ;D\n",
      "is upset that he can't update his Facebook by texting it... and might cry as a result  School today also. Blah!\n",
      "@Kenichan I dived many times for the ball. Managed to save 50%  The rest go out of bounds\n",
      "my whole body feels itchy and like its on fire \n",
      "@nationwideclass no, it's not behaving at all. i'm mad. why am i here? because I can't see you all over there. \n",
      "@Kwesidei not the whole crew \n",
      "Need a hug \n",
      "@LOLTrish hey  long time no see! Yes.. Rains a bit ,only a bit  LOL , I'm fine thanks , how's you ?\n",
      "@Tatiana_K nope they didn't have it \n",
      "@twittera que me muera ? \n"
     ]
    }
   ],
   "source": [
    "for t in df.tweet.head(10):\n",
    "    print(t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "717a54d8",
   "metadata": {
    "id": "717a54d8"
   },
   "outputs": [],
   "source": [
    "tweets_list = list(df.tweet)\n",
    "tweet_len = list(map(len, tweets_list))\n",
    "df[\"tweet_len\"] = tweet_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "73339b79",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "73339b79",
    "outputId": "ffba4f2b-0381-4628-ae6a-ae8217c0305d"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-2668b6c0-2363-490a-840c-e6e0d71f7fa2\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>target</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>datetime</th>\n",
       "      <th>query</th>\n",
       "      <th>id</th>\n",
       "      <th>tweet</th>\n",
       "      <th>tweet_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810369</td>\n",
       "      <td>Mon Apr 06 22:19:45 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>_TheSpecialOne_</td>\n",
       "      <td>@switchfoot http://twitpic.com/2y1zl - Awww, t...</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810672</td>\n",
       "      <td>Mon Apr 06 22:19:49 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>scotthamilton</td>\n",
       "      <td>is upset that he can't update his Facebook by ...</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1467810917</td>\n",
       "      <td>Mon Apr 06 22:19:53 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>mattycus</td>\n",
       "      <td>@Kenichan I dived many times for the ball. Man...</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811184</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>ElleCTF</td>\n",
       "      <td>my whole body feels itchy and like its on fire</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1467811193</td>\n",
       "      <td>Mon Apr 06 22:19:57 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>Karoli</td>\n",
       "      <td>@nationwideclass no, it's not behaving at all....</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2668b6c0-2363-490a-840c-e6e0d71f7fa2')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-2668b6c0-2363-490a-840c-e6e0d71f7fa2 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-2668b6c0-2363-490a-840c-e6e0d71f7fa2');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   target    tweet_id                      datetime     query  \\\n",
       "0       0  1467810369  Mon Apr 06 22:19:45 PDT 2009  NO_QUERY   \n",
       "1       0  1467810672  Mon Apr 06 22:19:49 PDT 2009  NO_QUERY   \n",
       "2       0  1467810917  Mon Apr 06 22:19:53 PDT 2009  NO_QUERY   \n",
       "3       0  1467811184  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY   \n",
       "4       0  1467811193  Mon Apr 06 22:19:57 PDT 2009  NO_QUERY   \n",
       "\n",
       "                id                                              tweet  \\\n",
       "0  _TheSpecialOne_  @switchfoot http://twitpic.com/2y1zl - Awww, t...   \n",
       "1    scotthamilton  is upset that he can't update his Facebook by ...   \n",
       "2         mattycus  @Kenichan I dived many times for the ball. Man...   \n",
       "3          ElleCTF    my whole body feels itchy and like its on fire    \n",
       "4           Karoli  @nationwideclass no, it's not behaving at all....   \n",
       "\n",
       "   tweet_len  \n",
       "0        115  \n",
       "1        111  \n",
       "2         89  \n",
       "3         47  \n",
       "4        111  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "fb64c768",
   "metadata": {
    "id": "fb64c768"
   },
   "outputs": [],
   "source": [
    "df = df[df['tweet'].apply(lambda x : x.isascii())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9891d7e9",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9891d7e9",
    "outputId": "63059bb4-a3ea-478d-bfc9-9a1591165f51"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['tweet'].apply(lambda x : x.isascii()).all()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e3d90e55",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 506
    },
    "id": "e3d90e55",
    "outputId": "ce18b4c7-8cba-41fc-c3e1-62f451149aa0"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 489,
       "width": 761
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(df[\"target\"]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "kTnMyJRmSuiv",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kTnMyJRmSuiv",
    "outputId": "09be0237-0815-45b9-cd1e-04c374761044"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1.])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(df.target / 4).unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "KZTZr-8FTJQS",
   "metadata": {
    "id": "KZTZr-8FTJQS"
   },
   "outputs": [],
   "source": [
    "df[\"target\"] = df.target / 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "CRwbGLhNTPNj",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "CRwbGLhNTPNj",
    "outputId": "d58f456c-45f5-42bd-dbb6-7b659866aaf3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1.])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.target.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "2GEMfBvbXyp2",
   "metadata": {
    "id": "2GEMfBvbXyp2"
   },
   "outputs": [],
   "source": [
    "df = df.sample(frac=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "db359060",
   "metadata": {
    "id": "db359060"
   },
   "outputs": [],
   "source": [
    "##################### Creating moc df for faster dvelepment, remove for the main run #####################\n",
    "df = df[:10000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "e48KvHjqX11j",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 506
    },
    "id": "e48KvHjqX11j",
    "outputId": "13c06778-1c81-4769-ed38-0f53fef839a8"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 489,
       "width": 745
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(df[\"target\"]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "47938e89",
   "metadata": {
    "id": "47938e89"
   },
   "outputs": [],
   "source": [
    "def remove_newlines_tabs(text):\n",
    "    formatted_text = text.replace('\\\\n', ' ').replace('\\n', ' ').replace('\\t',' ')\n",
    "    formatted_text = formatted_text.replace('\\\\', ' ').replace('. com', '.com')\n",
    "    return formatted_text\n",
    "\n",
    "def remove_mentions_hashtagSigns(text):\n",
    "    tokenized = text.split(\" \")\n",
    "    formatted_text = \"\"\n",
    "    for w in tokenized:\n",
    "        if not w:\n",
    "            continue\n",
    "        if w[0] == \"@\":\n",
    "            continue\n",
    "        if w[0] == \"#\":\n",
    "            formatted_text += f\"{w[1:]} \"\n",
    "        else:\n",
    "            formatted_text += f\"{w} \"\n",
    "            \n",
    "    return formatted_text.strip()\n",
    "\n",
    "def strip_html_tags(text):\n",
    "    soup = BeautifulSoup(text, \"html.parser\")\n",
    "    stripped_text = soup.get_text(separator=\" \")\n",
    "    return stripped_text\n",
    "\n",
    "def remove_punctuations(text):\n",
    "    formatted = \"\"\n",
    "    for char in text:\n",
    "        if char in string.punctuation:\n",
    "            continue\n",
    "        else:\n",
    "            formatted += char\n",
    "    return formatted\n",
    "\n",
    "def remove_links(text):\n",
    "    remove_https = re.sub(r'http\\S+', '', text)\n",
    "    remove_com = re.sub(r\"\\ [A-Za-z]*\\.com\", \" \", remove_https)\n",
    "    return remove_com\n",
    "\n",
    "def remove_whitespace(text):\n",
    "    pattern = re.compile(r'\\s+') \n",
    "    Without_whitespace = re.sub(pattern, ' ', text)\n",
    "    text = Without_whitespace.replace('?', ' ? ').replace(')', ') ')\n",
    "    return text\n",
    "\n",
    "def accented_characters_removal(text):\n",
    "    # Remove accented characters from text using unidecode.\n",
    "    # Unidecode() - It takes unicode data & tries to represent it to ASCII characters. \n",
    "    text = unidecode.unidecode(text)\n",
    "    return text\n",
    "\n",
    "def lower_casing_text(text):\n",
    "    return text.lower()\n",
    "\n",
    "def reducing_incorrect_character_repeatation(text):\n",
    "    Pattern_alpha = re.compile(r\"([A-Za-z])\\1{1,}\", re.DOTALL)\n",
    "    \n",
    "    # Limiting all the  repeatation to two characters.\n",
    "    Formatted_text = Pattern_alpha.sub(r\"\\1\\1\", text) \n",
    "    \n",
    "    # Pattern matching for all the punctuations that can occur\n",
    "    Pattern_Punct = re.compile(r'([.,/#!$%^&*?;:{}=_`~()+-])\\1{1,}')\n",
    "    \n",
    "    # Limiting punctuations in previously formatted string to only one.\n",
    "    Combined_Formatted = Pattern_Punct.sub(r'\\1', Formatted_text)\n",
    "    \n",
    "    # The below statement is replacing repeatation of spaces that occur more than two times with that of one occurrence.\n",
    "    Final_Formatted = re.sub(' {2,}',' ', Combined_Formatted)\n",
    "    return Final_Formatted\n",
    "\n",
    "\n",
    "def expand_contractions(text, contraction_mapping):\n",
    "    list_Of_tokens = text.split(' ')\n",
    "    for Word in list_Of_tokens: \n",
    "         if Word in contraction_mapping: \n",
    "            list_Of_tokens = [item.replace(Word, contraction_mapping[Word]) for item in list_Of_tokens]\n",
    "                \n",
    "    String_Of_tokens = ' '.join(str(e) for e in list_Of_tokens) \n",
    "    return String_Of_tokens\n",
    "\n",
    "def removing_special_characters(text):\n",
    "    formatted_text = re.sub(r\"[^a-zA-Z0-9:$-,%.?!]+\", ' ', text) \n",
    "    return formatted_text\n",
    "\n",
    "def remove_numbers(text):\n",
    "    formatted_text = re.sub(r\"[^a-zA-Z:$-,%.?!]+\", ' ', text)\n",
    "    return formatted_text\n",
    "\n",
    "stoplist = stopwords.words('english') \n",
    "stoplist = set(stoplist)\n",
    "def removing_stopwords(text):\n",
    "    text = repr(text)\n",
    "    No_StopWords = [word for word in word_tokenize(text) if word.lower() not in stoplist]\n",
    "    words_string = ' '.join(No_StopWords)    \n",
    "    return words_string\n",
    "\n",
    "def spelling_correction(text):\n",
    "    spell = Speller(lang='en')\n",
    "    Corrected_text = spell(text)\n",
    "    return Corrected_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "c3ecbe24",
   "metadata": {
    "id": "c3ecbe24"
   },
   "outputs": [],
   "source": [
    "with open('/content/gdrive/MyDrive/MIR-HW4/contraction_map.json') as f:\n",
    "    contraction_map = json.load(f)\n",
    "\n",
    "def cleaning_and_preprocessing(tweet,\n",
    "                              remove_newlines_=True,\n",
    "                              remove_mentions_hashtagSigns_=True,\n",
    "                              strip_html_tags_=True,\n",
    "                              remove_punctuations_=True,\n",
    "                              remove_links_=True,\n",
    "                              remove_whitespace_=True,\n",
    "                              accented_characters_removal_=True,\n",
    "                              lower_casing_text_=True,\n",
    "                              reducing_incorrect_character_repeatation_=True,\n",
    "                              expand_contractions_=True,\n",
    "                              remove_numbers_=True,\n",
    "                              removing_stopwords_=True,\n",
    "                              spelling_correction_=False):\n",
    "    \n",
    "    \"\"\"\n",
    "    input: a single text (e.g., a tweet or a sentence with type string).\n",
    "    output: a single text (e.g., a tweet or a sentence with type string) which is clean:)\n",
    "    \"\"\"\n",
    "    \n",
    "    ## Cleaning\n",
    "    text = tweet\n",
    "    if remove_newlines_:\n",
    "        text = remove_newlines_tabs(text)\n",
    "    if remove_mentions_hashtagSigns_:\n",
    "        text = remove_mentions_hashtagSigns(text)\n",
    "    if strip_html_tags_:\n",
    "        text = strip_html_tags(text)\n",
    "    if remove_punctuations_:\n",
    "        text = remove_punctuations(text)\n",
    "    if remove_links_:\n",
    "        text = remove_links(text)\n",
    "    if remove_whitespace_:\n",
    "        text = remove_whitespace(text)\n",
    "    if  accented_characters_removal_:   \n",
    "        text = accented_characters_removal(text)\n",
    "    if lower_casing_text_:\n",
    "        text = lower_casing_text(text)\n",
    "    if reducing_incorrect_character_repeatation_:\n",
    "        text = reducing_incorrect_character_repeatation(text)\n",
    "    if expand_contractions_:\n",
    "        text = expand_contractions(text, contraction_mapping=contraction_map)\n",
    "    if remove_numbers_:\n",
    "        text = remove_numbers(text)\n",
    "    \n",
    "    ## preprocessing\n",
    "    if removing_stopwords_:\n",
    "        text = removing_stopwords(text)\n",
    "    if spelling_correction_:\n",
    "        text = spelling_correction(text)\n",
    "    return text\n",
    "\n",
    "\n",
    "def lemmatization(text):\n",
    "    \"\"\"\n",
    "        input: a single string text (e.g., tweet)\n",
    "    \"\"\"\n",
    "    w_tokenizer = nltk.tokenize.WhitespaceTokenizer()\n",
    "    lemmatizer = nltk.stem.WordNetLemmatizer()\n",
    "    lemma = [lemmatizer.lemmatize(w,'v') for w in w_tokenizer.tokenize(text)]\n",
    "    return lemma    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5e184363",
   "metadata": {
    "id": "5e184363"
   },
   "outputs": [],
   "source": [
    "def preprocess_all_tweets(df):\n",
    "    \"\"\"\n",
    "        gets whole df as input.\n",
    "        returns df with new column called cleaned tweets.\n",
    "    \"\"\"\n",
    "    cleaned_tweets = []\n",
    "    for tweet in tqdm(df.tweet):\n",
    "        cleaned_tweet = cleaning_and_preprocessing(tweet)\n",
    "        cleaned_tweets.append(cleaned_tweet)\n",
    "    \n",
    "    df[\"cleaned_tweets\"] = cleaned_tweets\n",
    "    return df\n",
    "\n",
    "\n",
    "def lemmatize_all_tweets(df):\n",
    "    \"\"\"\n",
    "        gets whole df as input.\n",
    "        returns df with new column called lemmatized tweets.\n",
    "    \"\"\"\n",
    "    lemmatized_tweets = []\n",
    "    for tweet in tqdm(df.cleaned_tweets):\n",
    "        lemmatized_tweet = cleaning_and_preprocessing(tweet)\n",
    "        lemmatized_tweets.append(lemmatized_tweet)\n",
    "    \n",
    "    df[\"lemmatized_tweets\"] = lemmatized_tweets\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "6a2ba7d0",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6a2ba7d0",
    "outputId": "e5f1a3e3-80e1-45bf-960a-4dedd13914db"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:02<00:00, 3744.51it/s]\n"
     ]
    }
   ],
   "source": [
    "df = preprocess_all_tweets(df)\n",
    "# df = lemmatize_all_tweets(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "61fc58f7",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "61fc58f7",
    "outputId": "5562d22f-7e79-4f22-c156-115382dafe28"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-caa9c65a-48c2-4adb-880b-aa715b54da91\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>target</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>datetime</th>\n",
       "      <th>query</th>\n",
       "      <th>id</th>\n",
       "      <th>tweet</th>\n",
       "      <th>tweet_len</th>\n",
       "      <th>cleaned_tweets</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1575549</th>\n",
       "      <td>1.0</td>\n",
       "      <td>2189409732</td>\n",
       "      <td>Mon Jun 15 23:58:00 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>MooseUpdate</td>\n",
       "      <td>Cuddle snuggle beddy time!</td>\n",
       "      <td>27</td>\n",
       "      <td>'cuddle snuggle beddy time '</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211392</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1974478410</td>\n",
       "      <td>Sat May 30 13:13:02 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>slipintothis</td>\n",
       "      <td>Go away, depressing weather. You make me sad</td>\n",
       "      <td>45</td>\n",
       "      <td>'go away depressing weather make sad '</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1231067</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1991682769</td>\n",
       "      <td>Mon Jun 01 07:28:47 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>mschu22</td>\n",
       "      <td>My only day of nannying this week. 9-5.</td>\n",
       "      <td>40</td>\n",
       "      <td>'my day nannying week '</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>832974</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1557680607</td>\n",
       "      <td>Sun Apr 19 05:28:56 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>FrancisCollis</td>\n",
       "      <td>Sunday, sun is shining,  sitting in the garden...</td>\n",
       "      <td>96</td>\n",
       "      <td>'sunday sun shining sitting garden setting new...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1138833</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1976878335</td>\n",
       "      <td>Sat May 30 19:24:22 PDT 2009</td>\n",
       "      <td>NO_QUERY</td>\n",
       "      <td>sssemester</td>\n",
       "      <td>@RawHealer Haha, exactly!</td>\n",
       "      <td>26</td>\n",
       "      <td>'haha exactly '</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-caa9c65a-48c2-4adb-880b-aa715b54da91')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-caa9c65a-48c2-4adb-880b-aa715b54da91 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-caa9c65a-48c2-4adb-880b-aa715b54da91');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "         target    tweet_id                      datetime     query  \\\n",
       "1575549     1.0  2189409732  Mon Jun 15 23:58:00 PDT 2009  NO_QUERY   \n",
       "211392      0.0  1974478410  Sat May 30 13:13:02 PDT 2009  NO_QUERY   \n",
       "1231067     1.0  1991682769  Mon Jun 01 07:28:47 PDT 2009  NO_QUERY   \n",
       "832974      1.0  1557680607  Sun Apr 19 05:28:56 PDT 2009  NO_QUERY   \n",
       "1138833     1.0  1976878335  Sat May 30 19:24:22 PDT 2009  NO_QUERY   \n",
       "\n",
       "                    id                                              tweet  \\\n",
       "1575549    MooseUpdate                        Cuddle snuggle beddy time!    \n",
       "211392    slipintothis      Go away, depressing weather. You make me sad    \n",
       "1231067        mschu22           My only day of nannying this week. 9-5.    \n",
       "832974   FrancisCollis  Sunday, sun is shining,  sitting in the garden...   \n",
       "1138833     sssemester                         @RawHealer Haha, exactly!    \n",
       "\n",
       "         tweet_len                                     cleaned_tweets  \n",
       "1575549         27                       'cuddle snuggle beddy time '  \n",
       "211392          45             'go away depressing weather make sad '  \n",
       "1231067         40                            'my day nannying week '  \n",
       "832974          96  'sunday sun shining sitting garden setting new...  \n",
       "1138833         26                                    'haha exactly '  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ce5e6145",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ce5e6145",
    "outputId": "b1de15dd-81fb-478c-d14d-7a0d1cb5a22a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0.])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.target.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a922a1c",
   "metadata": {
    "id": "4a922a1c"
   },
   "source": [
    "# ----------- Embedding with Sent-Transformer, Classify with classic models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "-sBtsZKjirX-",
   "metadata": {
    "id": "-sBtsZKjirX-"
   },
   "outputs": [],
   "source": [
    "!pip install sentence_transformers\n",
    "from sentence_transformers import SentenceTransformer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "import sklearn.metrics as met\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "aa04d594",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aa04d594",
    "outputId": "2f415f5a-2d82-479b-b2c5-ee0989845c3d"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [01:11<00:00, 139.98it/s]\n"
     ]
    }
   ],
   "source": [
    "def create_embedding(df, model):\n",
    "    all_embeddings = []\n",
    "    for tweet in tqdm(df.cleaned_tweets):\n",
    "        \n",
    "        tweet_embedding = model.encode([tweet])\n",
    "        all_embeddings.append(tweet_embedding)\n",
    "\n",
    "    with open(\"/content/gdrive/MyDrive/all_embeddings.npy\", \"wb\") as f:\n",
    "        np.save(f, all_embeddings)\n",
    "    \n",
    "    return all_embeddings\n",
    "\n",
    "\n",
    "model = SentenceTransformer('all-MiniLM-L6-v2')\n",
    "embeddings = create_embedding(df, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "26f61b68",
   "metadata": {
    "id": "26f61b68"
   },
   "outputs": [],
   "source": [
    "df[\"embeddings\"] = embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "3bLLnd7kkduQ",
   "metadata": {
    "id": "3bLLnd7kkduQ"
   },
   "outputs": [],
   "source": [
    "y = df.pop(\"target\")\n",
    "data = df.embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "KFbWzAUUmdlG",
   "metadata": {
    "id": "KFbWzAUUmdlG"
   },
   "outputs": [],
   "source": [
    "dm = np.zeros((len(data), embeddings[0].shape[1]))\n",
    "\n",
    "for i, em in enumerate(embeddings):\n",
    "    dm[i, :] = em"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "T1Q1QMzuj9IN",
   "metadata": {
    "id": "T1Q1QMzuj9IN"
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(dm, y, test_size=0.15, shuffle=True, random_state=76)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "Ng5h_k7pmQ7i",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Ng5h_k7pmQ7i",
    "outputId": "20117c32-95f0-4f17-cf63-ea4bbfe4fb47"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((8500, 384), (8500,))"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "0LMa2Nz6kpaV",
   "metadata": {
    "id": "0LMa2Nz6kpaV"
   },
   "outputs": [],
   "source": [
    "\n",
    "rf_clf = RandomForestClassifier()\n",
    "rf_clf = rf_clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "JeuZ2AXkkqPg",
   "metadata": {
    "id": "JeuZ2AXkkqPg"
   },
   "outputs": [],
   "source": [
    "def score(y_test, pred):\n",
    "    if y_test.sum() == len(y_test) or y_test.sum() == 0: \n",
    "        return met.accuracy_score(y_test, pred), met.f1_score(y_test, pred),\\\n",
    "        met.precision_score(y_test, pred),\\\n",
    "        met.recall_score(y_test, pred), None\n",
    "    return met.accuracy_score(y_test, pred), met.f1_score(y_test, pred),\\\n",
    "    met.precision_score(y_test, pred),\\\n",
    "    met.recall_score(y_test, pred),\\\n",
    "    met.roc_auc_score(y_test, pred)\n",
    "\n",
    "def print_score(y_test, pred):\n",
    "    print('Accuracy:  ', met.accuracy_score(y_test, pred))\n",
    "    print('F1_score:  ', met.f1_score(y_test, pred))\n",
    "    print('Precision: ', met.precision_score(y_test, pred))\n",
    "    print('Recall:    ', met.recall_score(y_test, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "xt3zgJ-Omvr0",
   "metadata": {
    "id": "xt3zgJ-Omvr0"
   },
   "outputs": [],
   "source": [
    "rf_clf = RandomForestClassifier()\n",
    "rf_clf = rf_clf.fit(X_train, y_train)\n",
    "pred = rf_clf.predict(X_test)\n",
    "acc, f1, prec, rec, roc = score(y_test, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "opRY1Ejtm8xg",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "opRY1Ejtm8xg",
    "outputId": "e77c2fe4-a8ad-42b6-8888-6ea79e05dc3e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:   0.7086666666666667\n",
      "F1_score:   0.6937631394533987\n",
      "Precision:  0.7236842105263158\n",
      "Recall:     0.6662180349932705\n"
     ]
    }
   ],
   "source": [
    "print_score(y_test, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "7wjL0Jvmm9Fl",
   "metadata": {
    "id": "7wjL0Jvmm9Fl"
   },
   "outputs": [],
   "source": [
    "lg_clf = LogisticRegression()\n",
    "lg_clf = lg_clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "V9WY_ROgni6b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "V9WY_ROgni6b",
    "outputId": "70e451b5-2623-4cef-840f-e55006363533"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:   0.7386666666666667\n",
      "F1_score:   0.7315068493150685\n",
      "Precision:  0.7447698744769874\n",
      "Recall:     0.7187079407806191\n"
     ]
    }
   ],
   "source": [
    "lg_clf = LogisticRegression()\n",
    "lg_clf = lg_clf.fit(X_train, y_train)\n",
    "pred = lg_clf.predict(X_test)\n",
    "print_score(y_test, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "lbXNXpQVnq4c",
   "metadata": {
    "id": "lbXNXpQVnq4c"
   },
   "outputs": [],
   "source": [
    "svm_clf = LinearSVC()\n",
    "svm_clf = svm_clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "mCDY5-tcn6Qj",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mCDY5-tcn6Qj",
    "outputId": "6092c5f1-e491-4a6f-a6aa-406dc3eaf263"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:   0.734\n",
      "F1_score:   0.7283866575901975\n",
      "Precision:  0.7369146005509641\n",
      "Recall:     0.7200538358008075\n"
     ]
    }
   ],
   "source": [
    "pred = svm_clf.predict(X_test)\n",
    "print_score(y_test, pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "VZNeXujzJGH5",
   "metadata": {
    "id": "VZNeXujzJGH5"
   },
   "source": [
    "# ----------- Training deep BERT model for sentiment analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "418c2bbd",
   "metadata": {
    "id": "418c2bbd"
   },
   "outputs": [],
   "source": [
    "PRE_TRAINED_MODEL_NAME = 'bert-base-cased'\n",
    "tokenizer = BertTokenizer.from_pretrained(PRE_TRAINED_MODEL_NAME)\n",
    "\n",
    "configuration = BertConfig(num_attention_heads=3, num_hidden_layers=3, intermediate_size=1024)\n",
    "model = BertModel(configuration)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "hu63cAvPGxiY",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "hu63cAvPGxiY",
    "outputId": "6084eb78-baf8-4100-93f4-7ee0ab6c320a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Sentence: 'haha exactly '\n",
      "   Tokens: [\"'\", 'ha', '##ha', 'exactly', \"'\"]\n",
      "Token IDs: [112, 5871, 2328, 2839, 112]\n"
     ]
    }
   ],
   "source": [
    "sample_txt = df.cleaned_tweets.iloc[4]\n",
    "tokens = tokenizer.tokenize(sample_txt)\n",
    "token_ids = tokenizer.convert_tokens_to_ids(tokens)\n",
    "\n",
    "print(f' Sentence: {sample_txt}')\n",
    "print(f'   Tokens: {tokens}')\n",
    "print(f'Token IDs: {token_ids}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d482e87",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4d482e87",
    "outputId": "2bc2ffb9-df6a-4eb9-8acf-31d57e79eda1"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "dict_keys(['input_ids', 'attention_mask'])"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoding = tokenizer.encode_plus(\n",
    "    sample_txt,\n",
    "    max_length=32,\n",
    "    add_special_tokens=True, # Add '[CLS]' and '[SEP]'\n",
    "    return_token_type_ids=False,\n",
    "    pad_to_max_length=True,\n",
    "    return_attention_mask=True,\n",
    "    return_tensors='pt',  # Return PyTorch tensors\n",
    ")\n",
    "\n",
    "encoding.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "jV62Ryn8HMeI",
   "metadata": {
    "id": "jV62Ryn8HMeI"
   },
   "outputs": [],
   "source": [
    "class TweetsDataset(Dataset):\n",
    "\n",
    "  def __init__(self, tweets, targets, tokenizer, max_len):\n",
    "    self.tweets = tweets\n",
    "    self.targets = targets\n",
    "    self.tokenizer = tokenizer\n",
    "    self.max_len = max_len\n",
    "  \n",
    "  def __len__(self):\n",
    "    return len(self.tweets)\n",
    "  \n",
    "  def __getitem__(self, item):\n",
    "    tweet = str(self.tweets[item])\n",
    "    target = self.targets[item]\n",
    "\n",
    "    encoding = self.tokenizer.encode_plus(\n",
    "      tweet,\n",
    "      add_special_tokens=True,\n",
    "      max_length=self.max_len,\n",
    "      return_token_type_ids=False,\n",
    "      pad_to_max_length=True,\n",
    "      return_attention_mask=True,\n",
    "      return_tensors='pt',\n",
    "    )\n",
    "\n",
    "    return {\n",
    "      'tweet_text': tweet,\n",
    "      'input_ids': encoding['input_ids'].flatten(),\n",
    "      'attention_mask': encoding['attention_mask'].flatten(),\n",
    "      'targets': torch.tensor(target, dtype=torch.long)\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8hPBlz4_HORS",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8hPBlz4_HORS",
    "outputId": "f6ef8078-5652-4a77-f903-5d1221a61c13"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((160000, 8), (20000, 8), (20000, 8))"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train, df_test = train_test_split(df, test_size=0.2, random_state=RANDOM_SEED)\n",
    "df_val, df_test = train_test_split(df_test, test_size=0.5, random_state=RANDOM_SEED)\n",
    "\n",
    "df_train.shape, df_val.shape, df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "H9Zx88xdHiuJ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "H9Zx88xdHiuJ",
    "outputId": "d3c27af3-2a5e-4f91-945d-14aee72e4973"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['tweet_text', 'input_ids', 'attention_mask', 'targets'])"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def create_data_loader(df, tokenizer, max_len, batch_size):\n",
    "    ds = TweetsDataset(\n",
    "        tweets=df.cleaned_tweets.to_numpy(),\n",
    "        targets=df.target.to_numpy(),\n",
    "        tokenizer=tokenizer,\n",
    "        max_len=max_len)\n",
    "\n",
    "    return DataLoader(\n",
    "    ds,\n",
    "    batch_size=batch_size,\n",
    "    num_workers=4)\n",
    "\n",
    "\n",
    "BATCH_SIZE = 256\n",
    "MAX_LEN = 32\n",
    "\n",
    "train_data_loader = create_data_loader(df_train, tokenizer, MAX_LEN, BATCH_SIZE)\n",
    "val_data_loader = create_data_loader(df_val, tokenizer, MAX_LEN, BATCH_SIZE)\n",
    "test_data_loader = create_data_loader(df_test, tokenizer, MAX_LEN, BATCH_SIZE)\n",
    "\n",
    "data = next(iter(train_data_loader))\n",
    "data.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "nFbP3t93Ih2c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "nFbP3t93Ih2c",
    "outputId": "821ef9dc-b9b2-483f-f994-2f85bbfaaea3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([256, 32])\n",
      "torch.Size([256, 32])\n",
      "torch.Size([256])\n"
     ]
    }
   ],
   "source": [
    "print(data['input_ids'].shape)\n",
    "print(data['attention_mask'].shape)\n",
    "print(data['targets'].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "-txUDQ3HJ6KN",
   "metadata": {
    "id": "-txUDQ3HJ6KN"
   },
   "outputs": [],
   "source": [
    "# bert_model = BertModel.from_pretrained(PRE_TRAINED_MODEL_NAME)\n",
    "bert_model = model\n",
    "\n",
    "last_hidden_state, pooled_output = bert_model(\n",
    "  input_ids=encoding['input_ids'], \n",
    "  attention_mask=encoding['attention_mask'],\n",
    "  return_dict=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "qKVkBO24QZ2J",
   "metadata": {
    "id": "qKVkBO24QZ2J"
   },
   "outputs": [],
   "source": [
    "class SentimentClassifier(nn.Module):\n",
    "\n",
    "  def __init__(self, n_classes, bert_model):\n",
    "    super(SentimentClassifier, self).__init__()\n",
    "    self.bert = bert_model\n",
    "    self.drop = nn.Dropout(p=0.5)\n",
    "    self.out = nn.Linear(self.bert.config.hidden_size, n_classes)\n",
    "  \n",
    "  def forward(self, input_ids, attention_mask):\n",
    "    _, pooled_output = self.bert(\n",
    "      input_ids=input_ids,\n",
    "      attention_mask=attention_mask,\n",
    "      return_dict=False\n",
    "    )\n",
    "    output = self.drop(pooled_output)\n",
    "    return self.out(output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc46cabQRSz6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dc46cabQRSz6",
    "outputId": "7b548fdb-e8dd-4ba6-8b78-805be4463667"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([256, 32])\n",
      "torch.Size([256, 32])\n"
     ]
    }
   ],
   "source": [
    "model = SentimentClassifier(n_classes=len(df.target.unique()), bert_model=bert_model)\n",
    "model = model.to(device)\n",
    "\n",
    "input_ids = data['input_ids'].to(device)\n",
    "attention_mask = data['attention_mask'].to(device)\n",
    "\n",
    "print(input_ids.shape) # batch size x seq length\n",
    "print(attention_mask.shape) # batch size x seq length"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ohUXZyt8R4mE",
   "metadata": {
    "id": "ohUXZyt8R4mE"
   },
   "outputs": [],
   "source": [
    "EPOCHS = 10\n",
    "\n",
    "optimizer = AdamW(model.parameters(), lr=2e-5, correct_bias=False)\n",
    "total_steps = len(train_data_loader) * EPOCHS\n",
    "\n",
    "scheduler = get_linear_schedule_with_warmup(\n",
    "  optimizer,\n",
    "  num_warmup_steps=0,\n",
    "  num_training_steps=total_steps\n",
    ")\n",
    "\n",
    "loss_fn = nn.CrossEntropyLoss().to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "Se6jEt-PR5az",
   "metadata": {
    "id": "Se6jEt-PR5az"
   },
   "outputs": [],
   "source": [
    "def train_epoch(\n",
    "  model, \n",
    "  data_loader, \n",
    "  loss_fn, \n",
    "  optimizer, \n",
    "  device, \n",
    "  scheduler, \n",
    "  n_examples\n",
    "):\n",
    "  model = model.train()\n",
    "\n",
    "  losses = []\n",
    "  correct_predictions = 0\n",
    "  \n",
    "  for d in data_loader:\n",
    "    input_ids = d[\"input_ids\"].to(device)\n",
    "    attention_mask = d[\"attention_mask\"].to(device)\n",
    "    targets = d[\"targets\"].to(device)\n",
    "\n",
    "    outputs = model(\n",
    "      input_ids=input_ids,\n",
    "      attention_mask=attention_mask\n",
    "    )\n",
    "\n",
    "    _, preds = torch.max(outputs, dim=1)\n",
    "    loss = loss_fn(outputs, targets)\n",
    "\n",
    "    correct_predictions += torch.sum(preds == targets)\n",
    "    losses.append(loss.item())\n",
    "\n",
    "    loss.backward()\n",
    "    nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)\n",
    "    optimizer.step()\n",
    "    scheduler.step()\n",
    "    optimizer.zero_grad()\n",
    "\n",
    "  return correct_predictions.double() / n_examples, np.mean(losses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "CEfKw-2BR7ll",
   "metadata": {
    "id": "CEfKw-2BR7ll"
   },
   "outputs": [],
   "source": [
    "def eval_model(model, data_loader, loss_fn, device, n_examples):\n",
    "  model = model.to(device)\n",
    "  model = model.eval()\n",
    "\n",
    "  losses = []\n",
    "  correct_predictions = 0\n",
    "\n",
    "  with torch.no_grad():\n",
    "    for d in data_loader:\n",
    "      input_ids = d[\"input_ids\"].to(device)\n",
    "      attention_mask = d[\"attention_mask\"].to(device)\n",
    "      targets = d[\"targets\"].to(device)\n",
    "      \n",
    "\n",
    "      outputs = model(\n",
    "        input_ids=input_ids,\n",
    "        attention_mask=attention_mask\n",
    "      )\n",
    "\n",
    "      _, preds = torch.max(outputs, dim=1)\n",
    "      \n",
    "\n",
    "      loss = loss_fn(outputs, targets)\n",
    "\n",
    "      correct_predictions += torch.sum(preds == targets)\n",
    "      losses.append(loss.item())\n",
    "\n",
    "  return correct_predictions.double() / n_examples, np.mean(losses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "uHHjEWS-SCnN",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uHHjEWS-SCnN",
    "outputId": "6c611622-f076-4aad-b785-bb9934052fcc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "----------\n",
      "Train loss 0.542478877544403 accuracy 0.716325\n",
      "Val   loss 0.49829623404937456 accuracy 0.7566\n",
      "\n",
      "Epoch 2/10\n",
      "----------\n",
      "Train loss 0.47834808797836303 accuracy 0.77000625\n",
      "Val   loss 0.48865278272689144 accuracy 0.7624000000000001\n",
      "\n",
      "Epoch 3/10\n",
      "----------\n",
      "Train loss 0.4611141904354095 accuracy 0.7812375\n",
      "Val   loss 0.48937261293206036 accuracy 0.7635000000000001\n",
      "\n",
      "Epoch 4/10\n",
      "----------\n",
      "Train loss 0.448864683675766 accuracy 0.7891250000000001\n",
      "Val   loss 0.4930602704422383 accuracy 0.7641\n",
      "\n",
      "Epoch 5/10\n",
      "----------\n",
      "Train loss 0.4389632164478302 accuracy 0.79505\n",
      "Val   loss 0.49453792956810966 accuracy 0.7618\n",
      "\n",
      "Epoch 6/10\n",
      "----------\n",
      "Train loss 0.43057059297561645 accuracy 0.80025\n",
      "Val   loss 0.49850953040243706 accuracy 0.7624500000000001\n",
      "\n",
      "Epoch 7/10\n",
      "----------\n",
      "Train loss 0.42385883007049563 accuracy 0.80464375\n",
      "Val   loss 0.5031617521485195 accuracy 0.7615500000000001\n",
      "\n",
      "Epoch 8/10\n",
      "----------\n",
      "Train loss 0.41692423758506775 accuracy 0.80895625\n",
      "Val   loss 0.5083382371105726 accuracy 0.7615500000000001\n",
      "\n",
      "Epoch 9/10\n",
      "----------\n",
      "Train loss 0.41174539885520933 accuracy 0.81161875\n",
      "Val   loss 0.5110434116441992 accuracy 0.7597\n",
      "\n",
      "Epoch 10/10\n",
      "----------\n",
      "Train loss 0.40769476118087766 accuracy 0.81414375\n",
      "Val   loss 0.5141229222092447 accuracy 0.75885\n",
      "\n",
      "CPU times: user 18min 32s, sys: 10.3 s, total: 18min 43s\n",
      "Wall time: 20min 28s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "history = defaultdict(list)\n",
    "best_accuracy = 0\n",
    "\n",
    "for epoch in range(EPOCHS):\n",
    "\n",
    "  print(f'Epoch {epoch + 1}/{EPOCHS}')\n",
    "  print('-' * 10)\n",
    "\n",
    "  train_acc, train_loss = train_epoch(\n",
    "    model,\n",
    "    train_data_loader,    \n",
    "    loss_fn, \n",
    "    optimizer, \n",
    "    device, \n",
    "    scheduler, \n",
    "    len(df_train)\n",
    "  )\n",
    "\n",
    "  print(f'Train loss {train_loss} accuracy {train_acc}')\n",
    "\n",
    "  val_acc, val_loss = eval_model(\n",
    "    model,\n",
    "    val_data_loader,\n",
    "    loss_fn, \n",
    "    device, \n",
    "    len(df_val)\n",
    "  )\n",
    "\n",
    "  print(f'Val   loss {val_loss} accuracy {val_acc}')\n",
    "  print()\n",
    "\n",
    "  history['train_acc'].append(train_acc)\n",
    "  history['train_loss'].append(train_loss)\n",
    "  history['val_acc'].append(val_acc)\n",
    "  history['val_loss'].append(val_loss)\n",
    "\n",
    "  if val_acc > best_accuracy:\n",
    "    torch.save(model.state_dict(), 'best_model_state.bin')\n",
    "    best_accuracy = val_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "BJKWqWrBSIxd",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 523
    },
    "id": "BJKWqWrBSIxd",
    "outputId": "d0098078-99a6-4bae-c5f8-1abf1b8bcef2"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 506,
       "width": 732
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_acc = [history['train_acc'][i].item() for i in range(len(history[\"train_acc\"]))]\n",
    "val_acc = [history['val_acc'][i].item() for i in range(len(history[\"val_acc\"]))] \n",
    "\n",
    "plt.plot(train_acc, label='train accuracy')\n",
    "plt.plot(val_acc, label='validation accuracy')\n",
    "\n",
    "plt.title('Training history')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend()\n",
    "plt.ylim([0, 1]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "UFw7KhcJFmhv",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UFw7KhcJFmhv",
    "outputId": "9eb79957-c7b8-4fae-a2b4-4e2f72cb3419"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7643500000000001"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ldft = len(df_test)\n",
    "device = \"cuda\"\n",
    "\n",
    "test_acc, _ = eval_model(\n",
    "  model,\n",
    "  test_data_loader,\n",
    "  loss_fn,\n",
    "  device,\n",
    "  ldft,\n",
    ")\n",
    "\n",
    "test_acc.item()"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "HW4.ipynb",
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}