ML_base_modelling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DwIpNVJf5E5i"
   },
   "source": [
    "Building baseline ML model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1_AmuVe05PYB"
   },
   "source": [
    "Importing libraries and loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "zV02txmO46Wg",
    "outputId": "14877f2d-a5e7-4b63-9f2d-2eb4379adb5d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: emoji in c:\\python312\\lib\\site-packages (2.14.0)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 24.0 -> 24.3.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: contractions in c:\\python312\\lib\\site-packages (0.1.73)Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 24.0 -> 24.3.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Requirement already satisfied: textsearch>=0.0.21 in c:\\python312\\lib\\site-packages (from contractions) (0.0.24)\n",
      "Requirement already satisfied: anyascii in c:\\python312\\lib\\site-packages (from textsearch>=0.0.21->contractions) (0.3.2)\n",
      "Requirement already satisfied: pyahocorasick in c:\\python312\\lib\\site-packages (from textsearch>=0.0.21->contractions) (2.1.0)\n"
     ]
    }
   ],
   "source": [
    "# Installing additional libraries for text preprocessing\n",
    "%pip install emoji\n",
    "%pip install contractions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "yZbfrEFV46Wi"
   },
   "outputs": [],
   "source": [
    "# Data manipulation libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "from pprint import pprint\n",
    "\n",
    "# Text processing libraries\n",
    "import emoji\n",
    "import re\n",
    "import contractions\n",
    "import spacy\n",
    "from spacy.lang.en.stop_words import STOP_WORDS\n",
    "\n",
    "# Scikit-Learn packages\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.linear_model import RidgeClassifier\n",
    "from sklearn.multioutput import MultiOutputClassifier\n",
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "\n",
    "\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "2kwMnXLC46Wk",
    "outputId": "1934d34e-1cb7-412f-9f75-3d69ac3bac53"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2214, 6)\n",
      "(276, 7)\n",
      "(278, 7)\n"
     ]
    }
   ],
   "source": [
    "# Importing train, validation and test datasets with preprocessed texts and labels\n",
    "train_data = pd.read_csv(\"D:/Drive D/TU_Dresden/STUDY/LLM/semeval_project/semeval_project/data/train_data.csv\", index_col=False) \n",
    "val_data= pd.read_csv(\"D:/Drive D/TU_Dresden/STUDY/LLM/semeval_project/semeval_project/data/val_data.csv\", index_col=False)\n",
    "test_data= pd.read_csv(\"D:/Drive D/TU_Dresden/STUDY/LLM/semeval_project/semeval_project/data/test_data.csv\", index_col=False)\n",
    "\n",
    "# Shape validation\n",
    "print(train_data.shape)\n",
    "print(val_data.shape)\n",
    "print(test_data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wsTjJVOn5T-z"
   },
   "source": [
    "Additional preprocessings for basic Machine Learning tasks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "He1mZxLK46Wo"
   },
   "outputs": [],
   "source": [
    "# Additional text preprocessing\n",
    "train_data['clean_text'] = train_data['clean_text'].apply(lambda x: re.sub(r\"[^A-Za-z_]+\",\" \", x))\n",
    "test_data['clean_text'] = test_data['clean_text'].apply(lambda x: re.sub(r\"[^A-Za-z_]+\",\" \", x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "gFtySxqa46Wp",
    "outputId": "d3584a90-bea7-4e42-f440-e392dc8e4da7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[38;5;2m✔ Download and installation successful\u001b[0m\n",
      "You can now load the package via spacy.load('en_core_web_sm')\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 24.0 -> 24.3.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pip in c:\\python312\\lib\\site-packages (24.0)\n",
      "Collecting pip\n",
      "  Using cached pip-24.3.1-py3-none-any.whl.metadata (3.7 kB)\n",
      "Using cached pip-24.3.1-py3-none-any.whl (1.8 MB)\n",
      "Installing collected packages: pip\n",
      "  Attempting uninstall: pip\n",
      "    Found existing installation: pip 24.0\n",
      "    Uninstalling pip-24.0:\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'c:\\\\python312\\\\lib\\\\site-packages\\\\pip-24.0.dist-info\\\\AUTHORS.txt'\n",
      "Consider using the `--user` option or check the permissions.\n",
      "\n",
      "\n",
      "[notice] A new release of pip is available: 24.0 -> 24.3.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    }
   ],
   "source": [
    "# Download model\n",
    "!python -m spacy download en_core_web_sm -q\n",
    "\n",
    "%pip install --upgrade pip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "id": "bjTQeirm46Wq"
   },
   "outputs": [],
   "source": [
    "# Import English using en_core_web_sm.load()\n",
    "import spacy\n",
    "nlp = spacy.load(\"en_core_web_sm\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WUpNAFKy5gj6"
   },
   "source": [
    "Now creating a TF-IDF matrix that will help us represent each sample of our corpus using the importance and frequency of each word in the sample, but also in the whole corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "zngnSDTq46Wr"
   },
   "outputs": [],
   "source": [
    "# create my own model which works as a spaCy pipeline\n",
    "def spacy_tokenizer(sentence):\n",
    "    # Creating our token object, which is used to create documents with linguistic annotations.\n",
    "    mytokens = nlp(sentence)\n",
    "\n",
    "    # Lemmatizing each token and converting each token into lowercase\n",
    "    mytokens = [ word.lemma_.lower().strip() if word.lemma_ != \"-PRON-\" else word.lower_ for word in mytokens ]\n",
    "\n",
    "    # Removing stop words\n",
    "    mytokens = [ word for word in mytokens if word not in STOP_WORDS and word not in ['\\n', '\\n\\n'] ]\n",
    "\n",
    "    # return preprocessed list of tokens\n",
    "    return mytokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "id": "OtRo16dF46Wt"
   },
   "outputs": [],
   "source": [
    "# Creating tokenized documents\n",
    "tokenized_train_data = train_data[\"clean_text\"].apply(lambda desc: nlp(desc))\n",
    "tokenized_test_data = test_data[\"clean_text\"].apply(lambda desc: nlp(desc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "XYXBCtCh46Wt"
   },
   "outputs": [],
   "source": [
    "# Lemmatize each token and removing english stopwords\n",
    "tokenized_train_data = tokenized_train_data.apply(lambda x: [token.lemma_ for token in x if token.lemma_ not in STOP_WORDS])\n",
    "tokenized_test_data = tokenized_test_data.apply(lambda x: [token.lemma_ for token in x if token.lemma_ not in STOP_WORDS])\n",
    "\n",
    "# Creating clean data in our dataframes\n",
    "train_data[\"clean_token\"] = [\" \".join(x) for x in tokenized_train_data]\n",
    "test_data[\"clean_token\"] = [\" \".join(x) for x in tokenized_test_data]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5JJ3u1W55o-5"
   },
   "source": [
    "Create TF-IDF matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "PB7qBB0X46Wu",
    "outputId": "eea013d7-834d-4542-eba1-a3a6227d185f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2214, 1000)\n",
      "(278, 1000)\n"
     ]
    }
   ],
   "source": [
    "# TF-IDF vector with 1000 words vocabulary\n",
    "vectorizer = TfidfVectorizer(stop_words=\"english\", max_features=1000)\n",
    "\n",
    "# Fitting the vectorizer and transforming train and test data\n",
    "tfidf_train_data = vectorizer.fit_transform(train_data['clean_token'])\n",
    "tfidf_test_data = vectorizer.transform(test_data['clean_token'])\n",
    "\n",
    "# Transforming from generators to arrays\n",
    "tfidf_train_data = tfidf_train_data.toarray()\n",
    "tfidf_test_data = tfidf_test_data.toarray()\n",
    "\n",
    "# Validating the shape of train and test data\n",
    "print(tfidf_train_data.shape)\n",
    "print(tfidf_test_data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "h7GPiS355uFa"
   },
   "source": [
    "Train and test variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "id": "bW5oGb9946Wv",
    "outputId": "0d409bf1-b4b9-41a5-b1c8-80d5a0a79011"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The shape of X_train is :  (2214, 1000)\n",
      "The shape of y_train is :  (2214, 5)\n",
      "\n",
      "The shape of X_test is :  (278, 1000)\n",
      "The shape of y_test is :  (278, 5)\n"
     ]
    }
   ],
   "source": [
    "# Defining the emotion columns\n",
    "emotion = ['anger', 'fear', 'joy', 'sadness', 'surprise']\n",
    "\n",
    "# Defining train and test variables\n",
    "X_train =  tfidf_train_data\n",
    "y_train = train_data.loc[:,emotion].values\n",
    "\n",
    "X_test =  tfidf_test_data\n",
    "y_test = test_data.loc[:,emotion].values\n",
    "\n",
    "# Shape validation\n",
    "print(\"The shape of X_train is : \", X_train.shape)\n",
    "print(\"The shape of y_train is : \", y_train.shape)\n",
    "print()\n",
    "print(\"The shape of X_test is : \", X_test.shape)\n",
    "print(\"The shape of y_test is : \", y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LVRlY9nN5x0H"
   },
   "source": [
    "Simulating dummy predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "tWl1AUV246Wv",
    "outputId": "50c74eb7-3711-420b-e23c-41369b26f8cd"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>clean_text</th>\n",
       "      <th>anger</th>\n",
       "      <th>fear</th>\n",
       "      <th>joy</th>\n",
       "      <th>sadness</th>\n",
       "      <th>surprise</th>\n",
       "      <th>clean_token</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>but not very happy</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>well she is not gon na last the whole song lik...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>song like I audience torso pretty I use hand p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>she sat at her papa recliner sofa only to move...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>sit papa recliner sofa I start cling arm</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          clean_text  anger  fear  joy  \\\n",
       "0                                 but not very happy      0     0    1   \n",
       "1  well she is not gon na last the whole song lik...      0     0    1   \n",
       "2  she sat at her papa recliner sofa only to move...      0     0    0   \n",
       "\n",
       "   sadness  surprise                                        clean_token  \n",
       "0        1         0                                              happy  \n",
       "1        0         0  song like I audience torso pretty I use hand p...  \n",
       "2        0         0           sit papa recliner sofa I start cling arm  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Preview of data\n",
    "display(train_data.head(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "fGkK37La46Ww",
    "outputId": "62522ecb-0b6b-4bc8-ad10-6f41ac45a418"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, 0, 1],\n",
       "       [0, 0, 0, 0, 1],\n",
       "       [0, 0, 0, 0, 1],\n",
       "       ...,\n",
       "       [0, 0, 0, 0, 1],\n",
       "       [0, 0, 0, 0, 1],\n",
       "       [0, 0, 0, 0, 1]])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  predicting  emotion\n",
    "dummy_preds = np.zeros_like(y_test)\n",
    "dummy_preds[:,-1] = 1\n",
    "dummy_preds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UmoJs8qu52R3"
   },
   "source": [
    "Evaluation of dummy predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "tPnYDWk846Ww"
   },
   "outputs": [],
   "source": [
    "# Model evaluation function\n",
    "def model_eval(y_true, y_pred_labels, emotions):\n",
    "\n",
    "    # Defining variables\n",
    "    precision = []\n",
    "    recall = []\n",
    "    f1 = []\n",
    "\n",
    "    # Per emotion evaluation\n",
    "    idx2emotion = {i: e for i, e in enumerate(emotions)}\n",
    "\n",
    "    for i in range(len(emotions)):\n",
    "\n",
    "        # Computing precision, recall and f1-score\n",
    "        p, r, f1_score, _ = precision_recall_fscore_support(y_true[:, i], y_pred_labels[:, i], average=\"binary\")\n",
    "\n",
    "        # Append results in lists\n",
    "        precision.append(round(p, 2))\n",
    "        recall.append(round(r, 2))\n",
    "        f1.append(round(f1_score, 2))\n",
    "\n",
    "    # Macro evaluation\n",
    "    macro_p, macro_r, macro_f1_score, _ = precision_recall_fscore_support(y_true, y_pred_labels, average=\"macro\")\n",
    "\n",
    "    # Append results in lists\n",
    "    precision.append(round(macro_p, 2))\n",
    "    recall.append(round(macro_r, 2))\n",
    "    f1.append(round(macro_f1_score, 2))\n",
    "\n",
    "    # Converting results to a dataframe\n",
    "    df_results = pd.DataFrame({\"Precision\":precision, \"Recall\":recall, 'F1':f1})\n",
    "    df_results.index = emotions+['MACRO-AVERAGE']\n",
    "\n",
    "    return df_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "XUslLQQp46Ww",
    "outputId": "f1b529c0-cded-403a-8db1-adb1e698f71f"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>anger</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fear</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joy</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sadness</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>surprise</th>\n",
       "      <td>0.31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MACRO-AVERAGE</th>\n",
       "      <td>0.06</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.09</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Precision  Recall    F1\n",
       "anger               0.00     0.0  0.00\n",
       "fear                0.00     0.0  0.00\n",
       "joy                 0.00     0.0  0.00\n",
       "sadness             0.00     0.0  0.00\n",
       "surprise            0.31     1.0  0.47\n",
       "MACRO-AVERAGE       0.06     0.2  0.09"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Model evaluation\n",
    "model_eval(y_test, dummy_preds, emotion)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "egbRpvm656aS"
   },
   "source": [
    "Baseline model: Ridge Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "WNqmhClq46Wx",
    "outputId": "d1a51f95-da1d-4ca1-f4dd-2252fdedcdba"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-1 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-1 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-1 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-1 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-1 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>MultiOutputClassifier(estimator=RidgeClassifier(class_weight=&#x27;balanced&#x27;),\n",
       "                      n_jobs=-1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;MultiOutputClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.multioutput.MultiOutputClassifier.html\">?<span>Documentation for MultiOutputClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>MultiOutputClassifier(estimator=RidgeClassifier(class_weight=&#x27;balanced&#x27;),\n",
       "                      n_jobs=-1)</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">estimator: RidgeClassifier</label><div class=\"sk-toggleable__content fitted\"><pre>RidgeClassifier(class_weight=&#x27;balanced&#x27;)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;RidgeClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.linear_model.RidgeClassifier.html\">?<span>Documentation for RidgeClassifier</span></a></label><div class=\"sk-toggleable__content fitted\"><pre>RidgeClassifier(class_weight=&#x27;balanced&#x27;)</pre></div> </div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "MultiOutputClassifier(estimator=RidgeClassifier(class_weight='balanced'),\n",
       "                      n_jobs=-1)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Multi-label classification\n",
    "rc = RidgeClassifier(class_weight='balanced')\n",
    "classifier = MultiOutputClassifier(rc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "E1JFmwAr46Wx",
    "outputId": "a5bde8e4-523f-4b94-8fc1-ce3ae0a3b5d9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, 0, 0],\n",
       "       [0, 0, 1, 0, 1],\n",
       "       [1, 1, 0, 0, 1],\n",
       "       ...,\n",
       "       [0, 0, 1, 1, 0],\n",
       "       [0, 0, 0, 0, 0],\n",
       "       [0, 1, 0, 0, 0]])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Making predictions on Emotions\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "classifier_preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "IkWITmfJ46Wx",
    "outputId": "684a9a21-6097-4588-a010-12d5ed5cc7a2"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>anger</th>\n",
       "      <td>0.23</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fear</th>\n",
       "      <td>0.70</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joy</th>\n",
       "      <td>0.42</td>\n",
       "      <td>0.63</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sadness</th>\n",
       "      <td>0.47</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>surprise</th>\n",
       "      <td>0.49</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MACRO-AVERAGE</th>\n",
       "      <td>0.46</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Precision  Recall    F1\n",
       "anger               0.23    0.34  0.28\n",
       "fear                0.70    0.62  0.66\n",
       "joy                 0.42    0.63  0.50\n",
       "sadness             0.47    0.62  0.54\n",
       "surprise            0.49    0.62  0.54\n",
       "MACRO-AVERAGE       0.46    0.57  0.50"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Model evaluation\n",
    "model_eval(y_test, classifier_preds, emotion)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "id": "vgrVAd5E46Wy"
   },
   "outputs": [],
   "source": [
    "# Importing the required libraries\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate the prediction matrix in the plot\n",
    "def plot_confusion_matrix(y_true, y_pred, emotions):\n",
    "    \n",
    "    # Defining the confusion matrix\n",
    "    cm = confusion_matrix(y_true, y_pred) \n",
    "    \n",
    "    # Plotting the confusion matrix\n",
    "    plt.figure(figsize=(10, 10))\n",
    "    sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\", cbar=False, xticklabels=emotions, yticklabels=emotions)\n",
    "    plt.xlabel(\"Predicted Emotions\")\n",
    "    plt.ylabel(\"True Emotions\")\n",
    "    plt.title(\"Confusion Matrix\")\n",
    "\n",
    "# Plotting the confusion matrix\n",
    "plot_confusion_matrix(y_test.argmax(axis=1), classifier_preds.argmax(axis=1), emotion)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "id": "sCT0BZJ846Wy",
    "outputId": "d34e3de2-0c61-4e2d-8d17-4941a902af53"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>anger</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fear</th>\n",
       "      <td>0.64</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joy</th>\n",
       "      <td>0.80</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sadness</th>\n",
       "      <td>0.54</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>surprise</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MACRO-AVERAGE</th>\n",
       "      <td>0.55</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Precision  Recall    F1\n",
       "anger               0.00    0.00  0.00\n",
       "fear                0.64    0.84  0.72\n",
       "joy                 0.80    0.13  0.22\n",
       "sadness             0.54    0.18  0.27\n",
       "surprise            0.75    0.17  0.28\n",
       "MACRO-AVERAGE       0.55    0.26  0.30"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Similarly use the Logistic Regression classifier for multi-label classification\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "lc = LogisticRegression(max_iter=1000)\n",
    "classifier = MultiOutputClassifier(lc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "\n",
    "# Making predictions on Emotions\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "classifier_preds\n",
    "\n",
    "# Model evaluation\n",
    "model_eval(y_test, classifier_preds, emotion)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Checking Logistic regression model accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The model accuracy is:  0.21942446043165467\n"
     ]
    }
   ],
   "source": [
    "#calculate the model accuracy\n",
    "accuracy = accuracy_score(y_test, classifier_preds)\n",
    "print(\"The model accuracy is: \", accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>anger</th>\n",
       "      <td>0.40</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fear</th>\n",
       "      <td>0.66</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joy</th>\n",
       "      <td>0.58</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sadness</th>\n",
       "      <td>0.48</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>surprise</th>\n",
       "      <td>0.54</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MACRO-AVERAGE</th>\n",
       "      <td>0.53</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Precision  Recall    F1\n",
       "anger               0.40    0.11  0.17\n",
       "fear                0.66    0.75  0.71\n",
       "joy                 0.58    0.34  0.43\n",
       "sadness             0.48    0.29  0.36\n",
       "surprise            0.54    0.30  0.39\n",
       "MACRO-AVERAGE       0.53    0.36  0.41"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Similarly use the Random Forest classifier for multi-label classification\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "classifier = MultiOutputClassifier(rf, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "\n",
    "# Making predictions on Emotions\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "classifier_preds\n",
    "\n",
    "# Model evaluation\n",
    "model_eval(y_test, classifier_preds, emotion)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Checking Random Forest model accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The model accuracy is:  0.2302158273381295\n"
     ]
    }
   ],
   "source": [
    "# #calculate the model accuracy\n",
    "accuracy = accuracy_score(y_test, classifier_preds)\n",
    "print(\"The model accuracy is: \", accuracy)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ridge Classifier\n",
      "               Precision  Recall    F1\n",
      "anger               0.23    0.34  0.28\n",
      "fear                0.70    0.62  0.66\n",
      "joy                 0.42    0.63  0.50\n",
      "sadness             0.47    0.62  0.54\n",
      "surprise            0.49    0.62  0.54\n",
      "MACRO-AVERAGE       0.46    0.57  0.50\n",
      "\n",
      "Logistic Regression\n",
      "               Precision  Recall    F1\n",
      "anger               0.00    0.00  0.00\n",
      "fear                0.64    0.84  0.72\n",
      "joy                 0.80    0.13  0.22\n",
      "sadness             0.54    0.18  0.27\n",
      "surprise            0.75    0.17  0.28\n",
      "MACRO-AVERAGE       0.55    0.26  0.30\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest\n",
      "               Precision  Recall    F1\n",
      "anger               0.40    0.11  0.17\n",
      "fear                0.66    0.75  0.71\n",
      "joy                 0.58    0.34  0.43\n",
      "sadness             0.48    0.29  0.36\n",
      "surprise            0.54    0.30  0.39\n",
      "MACRO-AVERAGE       0.53    0.36  0.41\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# comparision of the models using the model evaluation function\n",
    "# Ridge Classifier\n",
    "rc = RidgeClassifier(class_weight='balanced')\n",
    "classifier = MultiOutputClassifier(rc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "print(\"Ridge Classifier\")\n",
    "print(model_eval(y_test, classifier_preds, emotion))\n",
    "print()\n",
    "\n",
    "# Logistic Regression\n",
    "lc = LogisticRegression(max_iter=1000)\n",
    "classifier = MultiOutputClassifier(lc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "print(\"Logistic Regression\")\n",
    "print(model_eval(y_test, classifier_preds, emotion))\n",
    "print()\n",
    "\n",
    "# Random Forest\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "classifier = MultiOutputClassifier(rf, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "print(\"Random Forest\")\n",
    "print(model_eval(y_test, classifier_preds, emotion))\n",
    "print()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Prerana\\AppData\\Roaming\\Python\\Python312\\site-packages\\sklearn\\metrics\\_classification.py:1531: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1cAAAHDCAYAAADIquCMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8ekN5oAAAACXBIWXMAAA9hAAAPYQGoP6dpAABCHElEQVR4nO3deVwVVePH8S8XZXUXZTECTUUtBFcyNc1QXFNzzxLJJVNTo6w0E7fEXJBM07JcKhceLZcnl1zSLDMtcd9yfVzBLUFJ0WB+f/Tj5hVc0NEr8Xm/XvOSe+bMmXNG516/zMy5DoZhGAIAAAAA3BOLvTsAAAAAAP8GhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwDAA+Hg4KAhQ4Zke7sjR47IwcFBM2bMML1P/0b+/v7q3LmzvbsBALkS4QoAcpEZM2bIwcFBDg4O+umnnzKtNwxDvr6+cnBwUNOmTe3Qw3uXmJioN998U+XKlZObm5vc3d1VpUoVjRgxQhcuXLB39wAA/2J57N0BAMCD5+LiotmzZ6tWrVo25T/88IOOHz8uZ2dnO/Xs3vz6669q3LixLl26pBdffFFVqlSRJP32228aNWqU1q1bpxUrVti5l/fXvn37ZLHwu1MAsAfCFQDkQo0bN9a8efM0YcIE5cnzz0fB7NmzVaVKFZ09e9aOvbs7Fy5cUMuWLeXo6KgtW7aoXLlyNuvff/99TZ061U69u78Mw9CVK1fk6uqaY4MxAPwb8KstAMiFOnTooHPnzmnlypXWsqtXr2r+/Pl64YUXstwmJSVFb7zxhnx9feXs7KyAgACNHTtWhmHY1EtNTdXrr7+uYsWKKX/+/Hruued0/PjxLNs8ceKEXn75ZXl6esrZ2VmPP/64pk2bdldj+uSTT3TixAnFxMRkClaS5OnpqUGDBtmUffzxx3r88cfl7OwsHx8f9erVK9Otg3Xr1tUTTzyh7du3q06dOnJzc1Pp0qU1f/58SX9f7QsJCZGrq6sCAgK0atUqm+2HDBkiBwcH7d27V23btlWBAgVUtGhR9e3bV1euXLGpO336dNWrV0/FixeXs7OzKlSooMmTJ2cai7+/v5o2barvvvtOVatWlaurqz755BPruuufubp27ZqGDh2qMmXKyMXFRUWLFlWtWrVs/u4l6fvvv1ft2rXl7u6uQoUKqXnz5tqzZ0+WYzlw4IA6d+6sQoUKqWDBgoqIiNCff/6Zxd8KAOQuhCsAyIX8/f1Vo0YNzZkzx1q2bNkyJSUlqX379pnqG4ah5557TuPHj1fDhg0VExOjgIAA9e/fX5GRkTZ1u3btqtjYWDVo0ECjRo1S3rx51aRJk0xtJiYm6sknn9SqVavUu3dvffjhhypdurS6dOmi2NjYbI9p8eLFcnV1VevWre+o/pAhQ9SrVy/5+Pho3LhxatWqlT755BM1aNBA165ds6n7xx9/qGnTpgoJCdHo0aPl7Oys9u3bKy4uTu3bt1fjxo01atQopaSkqHXr1rp48WKm/bVt21ZXrlxRdHS0GjdurAkTJqh79+42dSZPniw/Pz8NHDhQ48aNk6+vr3r27KlJkyZlam/fvn3q0KGD6tevrw8//FDBwcE3HefQoUP1zDPPaOLEiXr33Xf16KOPKj4+3lpn1apVCgsL0+nTpzVkyBBFRkbq559/Vs2aNXXkyJEsx3Lx4kVFR0erbdu2mjFjhoYOHXoHRx0A/uUMAECuMX36dEOS8euvvxoTJ0408ufPb/z555+GYRhGmzZtjGeeecYwDMPw8/MzmjRpYt1u4cKFhiRjxIgRNu21bt3acHBwMA4cOGAYhmFs3brVkGT07NnTpt4LL7xgSDKioqKsZV26dDG8vb2Ns2fP2tRt3769UbBgQWu/Dh8+bEgypk+ffsuxFS5c2AgKCrqj43D69GnDycnJaNCggZGWlmYtnzhxoiHJmDZtmrWsTp06hiRj9uzZ1rK9e/cakgyLxWL88ssv1vLvvvsuU1+joqIMScZzzz1n04eePXsakoxt27ZZyzLGfL2wsDCjVKlSNmV+fn6GJGP58uWZ6vv5+Rnh4eHW10FBQTZ/l1kJDg42ihcvbpw7d85atm3bNsNisRidOnXKNJaXX37ZZvuWLVsaRYsWveU+ACA34MoVAORSbdu21eXLl/Xtt9/q4sWL+vbbb296S+DSpUvl6OioPn362JS/8cYbMgxDy5Yts9aTlKlev379bF4bhqGvv/5azZo1k2EYOnv2rHUJCwtTUlKSzZWVO5GcnKz8+fPfUd1Vq1bp6tWr6tevn83kD926dVOBAgW0ZMkSm/r58uWzuaIXEBCgQoUKqXz58goJCbGWZ/x86NChTPvs1auXzevXXntN0j/HTJJcXV2tPyclJens2bOqU6eODh06pKSkJJvtS5YsqbCwsNuOtVChQtq1a5f279+f5fpTp05p69at6ty5s4oUKWItr1ixourXr2/Tvww9evSweV27dm2dO3dOycnJt+0PAPybEa4AIJcqVqyYQkNDNXv2bH3zzTdKS0u76S11//vf/+Tj45MpvJQvX966PuNPi8Wixx57zKZeQECAzeszZ87owoUL+vTTT1WsWDGbJSIiQpJ0+vTpbI2nQIECWd6Od7PxZNUvJycnlSpVyro+wyOPPCIHBwebsoIFC8rX1zdTmfT3bYQ3KlOmjM3rxx57TBaLxea2u/Xr1ys0NNT63FOxYsU0cOBAScoyXN2JYcOG6cKFCypbtqwCAwPVv39/bd++3br+ZsdC+vvv9+zZs0pJSbEpf/TRR21eFy5cWFLW4waA3ITZAgEgF3vhhRfUrVs3JSQkqFGjRipUqNAD2W96erok6cUXX1R4eHiWdSpWrJitNsuVK6etW7fq6tWrcnJyuuc+Xs/R0TFb5cYNk3xk5cawdvDgQT377LMqV66cYmJi5OvrKycnJy1dulTjx4+3HrMM11/lupWnn35aBw8e1KJFi7RixQp99tlnGj9+vKZMmaKuXbveURs3updxA8C/GVeuACAXa9mypSwWi3755Zeb3hIoSX5+fjp58mSmK0N79+61rs/4Mz09XQcPHrSpt2/fPpvXGTMJpqWlKTQ0NMulePHi2RpLs2bNdPnyZX399de3rZvR3xv7dfXqVR0+fNi63kw33pZ34MABpaeny9/fX5L03//+V6mpqVq8eLFeeeUVNW7cWKGhoXccom6lSJEiioiI0Jw5c3Ts2DFVrFhRQ4YMkXTzYyH9/ffr4eEhd3f3e+4DAOQGhCsAyMXy5cunyZMna8iQIWrWrNlN6zVu3FhpaWmaOHGiTfn48ePl4OCgRo0aSZL1zwkTJtjUu3H2P0dHR7Vq1Upff/21du7cmWl/Z86cyfZYevToIW9vb73xxhv6/fffM60/ffq0RowYIUkKDQ2Vk5OTJkyYYHO15fPPP1dSUlKWsxveqxtn/Pvoo48k/XPMMq4GXd+fpKQkTZ8+/Z72e+7cOZvX+fLlU+nSpZWamipJ8vb2VnBwsGbOnGkzDf3OnTu1YsUKNW7c+J72DwC5CbcFAkAud7Pb8q7XrFkzPfPMM3r33Xd15MgRBQUFacWKFVq0aJH69etnfcYqODhYHTp00Mcff6ykpCQ99dRTWr16tQ4cOJCpzVGjRmnNmjUKCQlRt27dVKFCBZ0/f17x8fFatWqVzp8/n61xFC5cWAsWLFDjxo0VHBysF198UVWqVJEkxcfHa86cOapRo4akv6+cDRgwQEOHDlXDhg313HPPad++ffr4449VrVo1vfjii9na9504fPiwnnvuOTVs2FAbNmzQV199pRdeeEFBQUGSpAYNGsjJyUnNmjXTK6+8okuXLmnq1KkqXry4Tp06ddf7rVChgurWrasqVaqoSJEi+u233zR//nz17t3bWmfMmDFq1KiRatSooS5duujy5cv66KOPVLBgQesVLgDA7RGuAAC3ZbFYtHjxYg0ePFhxcXGaPn26/P39NWbMGL3xxhs2dadNm6ZixYpp1qxZWrhwoerVq6clS5ZkmvzB09NTmzZt0rBhw/TNN9/o448/VtGiRfX444/rgw8+uKt+hoSEaOfOnRozZoyWLFmiL7/8UhaLReXLl9c777xjEyiGDBmiYsWKaeLEiXr99ddVpEgRde/eXSNHjlTevHnvav+3EhcXp8GDB+udd95Rnjx51Lt3b40ZM8a6PiAgQPPnz9egQYP05ptvysvLS6+++qqKFSuml19++a7326dPHy1evFgrVqxQamqq/Pz8NGLECPXv399aJzQ0VMuXL1dUVJQGDx6svHnzqk6dOvrggw/ueOIMAIDkYPD0KQAA903Gl/ieOXNGHh4e9u4OAOA+4pkrAAAAADAB4QoAAAAATEC4AgAAAAAT8MwVAAAAAJiAK1cAAAAAYALCFQAAAACYgO+5ykJ6erpOnjyp/Pnzy8HBwd7dAQAAAGAnhmHo4sWL8vHxkcVy62tThKssnDx5MtOXXQIAAADIvY4dO6ZHHnnklnUIV1nInz+/pL8PYIECBezcGwAAAAD2kpycLF9fX2tGuBXCVRYybgUsUKAA4QoAAADAHT0uxIQWAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAny2LsDAAAAuZH/O0vs3QXgoXZkVBN7dyHbuHIFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmCCPvTsAAPiH/ztL7N0F4KF2ZFQTe3cBAG6KK1cAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYIKHIlxNmjRJ/v7+cnFxUUhIiDZt2nTTulOnTlXt2rVVuHBhFS5cWKGhoZnqG4ahwYMHy9vbW66urgoNDdX+/fvv9zAAAAAA5GJ2D1dxcXGKjIxUVFSU4uPjFRQUpLCwMJ0+fTrL+mvXrlWHDh20Zs0abdiwQb6+vmrQoIFOnDhhrTN69GhNmDBBU6ZM0caNG+Xu7q6wsDBduXLlQQ0LAAAAQC5j93AVExOjbt26KSIiQhUqVNCUKVPk5uamadOmZVl/1qxZ6tmzp4KDg1WuXDl99tlnSk9P1+rVqyX9fdUqNjZWgwYNUvPmzVWxYkV98cUXOnnypBYuXPgARwYAAAAgN7FruLp69ao2b96s0NBQa5nFYlFoaKg2bNhwR238+eefunbtmooUKSJJOnz4sBISEmzaLFiwoEJCQu64TQAAAADIrjz23PnZs2eVlpYmT09Pm3JPT0/t3bv3jtp4++235ePjYw1TCQkJ1jZubDNj3Y1SU1OVmppqfZ2cnHzHYwAAAAAA6SG4LfBejBo1SnPnztWCBQvk4uJy1+1ER0erYMGC1sXX19fEXgIAAADIDewarjw8POTo6KjExESb8sTERHl5ed1y27Fjx2rUqFFasWKFKlasaC3P2C47bQ4YMEBJSUnW5dixY3czHAAAAAC5mF3DlZOTk6pUqWKdjEKSdXKKGjVq3HS70aNHa/jw4Vq+fLmqVq1qs65kyZLy8vKyaTM5OVkbN268aZvOzs4qUKCAzQIAAAAA2WHXZ64kKTIyUuHh4apataqqV6+u2NhYpaSkKCIiQpLUqVMnlShRQtHR0ZKkDz74QIMHD9bs2bPl7+9vfY4qX758ypcvnxwcHNSvXz+NGDFCZcqUUcmSJfXee+/Jx8dHLVq0sNcwAQAAAPzL2T1ctWvXTmfOnNHgwYOVkJCg4OBgLV++3DohxdGjR2Wx/HOBbfLkybp69apat25t005UVJSGDBkiSXrrrbeUkpKi7t2768KFC6pVq5aWL19+T89lAQAAAMCtOBiGYdi7Ew+b5ORkFSxYUElJSdwiCOCB8n9nib27ADzUjoxqYu8umIbzHbi1h+V8z042yNGzBQIAAADAw4JwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYII89u4Abs//nSX27gLw0Dsyqom9uwAAAHI5rlwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmsHu4mjRpkvz9/eXi4qKQkBBt2rTppnV37dqlVq1ayd/fXw4ODoqNjc1UZ8iQIXJwcLBZypUrdx9HAAAAAAB2DldxcXGKjIxUVFSU4uPjFRQUpLCwMJ0+fTrL+n/++adKlSqlUaNGycvL66btPv744zp16pR1+emnn+7XEAAAAABAkp3DVUxMjLp166aIiAhVqFBBU6ZMkZubm6ZNm5Zl/WrVqmnMmDFq3769nJ2db9punjx55OXlZV08PDzu1xAAAAAAQJIdw9XVq1e1efNmhYaG/tMZi0WhoaHasGHDPbW9f/9++fj4qFSpUurYsaOOHj16y/qpqalKTk62WQAAAAAgO+wWrs6ePau0tDR5enralHt6eiohIeGu2w0JCdGMGTO0fPlyTZ48WYcPH1bt2rV18eLFm24THR2tggULWhdfX9+73j8AAACA3MnuE1qYrVGjRmrTpo0qVqyosLAwLV26VBcuXNB//vOfm24zYMAAJSUlWZdjx449wB4DAAAA+DfIY68de3h4yNHRUYmJiTbliYmJt5ysIrsKFSqksmXL6sCBAzet4+zsfMtnuAAAAADgdux25crJyUlVqlTR6tWrrWXp6elavXq1atSoYdp+Ll26pIMHD8rb29u0NgEAAADgRna7ciVJkZGRCg8PV9WqVVW9enXFxsYqJSVFERERkqROnTqpRIkSio6OlvT3JBi7d++2/nzixAlt3bpV+fLlU+nSpSVJb775ppo1ayY/Pz+dPHlSUVFRcnR0VIcOHewzSAAAAAC5gl3DVbt27XTmzBkNHjxYCQkJCg4O1vLly62TXBw9elQWyz8X106ePKlKlSpZX48dO1Zjx45VnTp1tHbtWknS8ePH1aFDB507d07FihVTrVq19Msvv6hYsWIPdGwAAAAAche7hitJ6t27t3r37p3luozAlMHf31+GYdyyvblz55rVNQAAAAC4Y/+62QIBAAAAwB4IVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmCDb4crf31/Dhg3T0aNH70d/AAAAACBHyna46tevn7755huVKlVK9evX19y5c5Wamno/+gYAAAAAOcZdhautW7dq06ZNKl++vF577TV5e3urd+/eio+Pvx99BAAAAICH3l0/c1W5cmVNmDBBJ0+eVFRUlD777DNVq1ZNwcHBmjZtmgzDMLOfAAAAAPBQy3O3G167dk0LFizQ9OnTtXLlSj355JPq0qWLjh8/roEDB2rVqlWaPXu2mX0FAAAAgIdWtsNVfHy8pk+frjlz5shisahTp04aP368ypUrZ63TsmVLVatWzdSOAgAAAMDDLNvhqlq1aqpfv74mT56sFi1aKG/evJnqlCxZUu3btzelgwAAAACQE2Q7XB06dEh+fn63rOPu7q7p06ffdacAAAAAIKfJ9oQWp0+f1saNGzOVb9y4Ub/99pspnQIAAACAnCbb4apXr146duxYpvITJ06oV69epnQKAAAAAHKabIer3bt3q3LlypnKK1WqpN27d5vSKQAAAADIabIdrpydnZWYmJip/NSpU8qT565ndgcAAACAHC3b4apBgwYaMGCAkpKSrGUXLlzQwIEDVb9+fVM7BwAAAAA5RbYvNY0dO1ZPP/20/Pz8VKlSJUnS1q1b5enpqS+//NL0DgIAAABATpDtcFWiRAlt375ds2bN0rZt2+Tq6qqIiAh16NAhy++8AgAAAIDc4K4eknJ3d1f37t3N7gsAAAAA5Fh3PQPF7t27dfToUV29etWm/LnnnrvnTgEAAABATpPtcHXo0CG1bNlSO3bskIODgwzDkCQ5ODhIktLS0sztIQAAAADkANmeLbBv374qWbKkTp8+LTc3N+3atUvr1q1T1apVtXbt2vvQRQAAAAB4+GX7ytWGDRv0/fffy8PDQxaLRRaLRbVq1VJ0dLT69OmjLVu23I9+AgAAAMBDLdtXrtLS0pQ/f35JkoeHh06ePClJ8vPz0759+8ztHQAAAADkENm+cvXEE09o27ZtKlmypEJCQjR69Gg5OTnp008/ValSpe5HHwEAAADgoZftcDVo0CClpKRIkoYNG6amTZuqdu3aKlq0qOLi4kzvIAAAAADkBNkOV2FhYdafS5curb179+r8+fMqXLiwdcZAAAAAAMhtsvXM1bVr15QnTx7t3LnTprxIkSIEKwAAAAC5WrbCVd68efXoo4/yXVYAAAAAcINszxb47rvvauDAgTp//vz96A8AAAAA5EjZfuZq4sSJOnDggHx8fOTn5yd3d3eb9fHx8aZ1DgAAAAByimyHqxYtWtyHbgAAAABAzpbtcBUVFXU/+gEAAAAAOVq2n7kCAAAAAGSW7StXFovlltOuM5MgAAAAgNwo2+FqwYIFNq+vXbumLVu2aObMmRo6dKhpHQMAAACAnCTb4ap58+aZylq3bq3HH39ccXFx6tKliykdAwAAAICcxLRnrp588kmtXr3arOYAAAAAIEcxJVxdvnxZEyZMUIkSJcxoDgAAAABynGzfFli4cGGbCS0Mw9DFixfl5uamr776ytTOAQAAAEBOke1wNX78eJtwZbFYVKxYMYWEhKhw4cKmdg4AAAAAcopsh6vOnTvfh24AAAAAQM6W7Weupk+frnnz5mUqnzdvnmbOnGlKpwAAAAAgp8l2uIqOjpaHh0em8uLFi2vkyJGmdAoAAAAAcppsh6ujR4+qZMmSmcr9/Px09OhRUzoFAAAAADlNtsNV8eLFtX379kzl27ZtU9GiRU3pFAAAAADkNNkOVx06dFCfPn20Zs0apaWlKS0tTd9//7369u2r9u3b348+AgAAAMBDL9uzBQ4fPlxHjhzRs88+qzx5/t48PT1dnTp14pkrAAAAALlWtsOVk5OT4uLiNGLECG3dulWurq4KDAyUn5/f/egfAAAAAOQI2Q5XGcqUKaMyZcqY2RcAAAAAyLGy/cxVq1at9MEHH2QqHz16tNq0aWNKpwAAAAAgp8l2uFq3bp0aN26cqbxRo0Zat26dKZ0CAAAAgJwm2+Hq0qVLcnJyylSeN29eJScnm9IpAAAAAMhpsh2uAgMDFRcXl6l87ty5qlChgimdAgAAAICcJtsTWrz33nt6/vnndfDgQdWrV0+StHr1as2ePVvz5883vYMAAAAAkBNkO1w1a9ZMCxcu1MiRIzV//ny5uroqKChI33//vYoUKXI/+ggAAAAAD727moq9SZMmatKkiSQpOTlZc+bM0ZtvvqnNmzcrLS3N1A4CAAAAQE6Q7WeuMqxbt07h4eHy8fHRuHHjVK9ePf3yyy9m9g0AAAAAcoxsXblKSEjQjBkz9Pnnnys5OVlt27ZVamqqFi5cyGQWAAAAAHK1O75y1axZMwUEBGj79u2KjY3VyZMn9dFHH93PvgEAAABAjnHHV66WLVumPn366NVXX1WZMmXuZ58AAAAAIMe54ytXP/30ky5evKgqVaooJCREEydO1NmzZ++5A5MmTZK/v79cXFwUEhKiTZs23bTurl271KpVK/n7+8vBwUGxsbH33CYAAAAAmOGOw9WTTz6pqVOn6tSpU3rllVc0d+5c+fj4KD09XStXrtTFixezvfO4uDhFRkYqKipK8fHxCgoKUlhYmE6fPp1l/T///FOlSpXSqFGj5OXlZUqbAAAAAGCGbM8W6O7urpdfflk//fSTduzYoTfeeEOjRo1S8eLF9dxzz2WrrZiYGHXr1k0RERGqUKGCpkyZIjc3N02bNi3L+tWqVdOYMWPUvn17OTs7m9ImAAAAAJjhrqdil6SAgACNHj1ax48f15w5c7K17dWrV7V582aFhob+0xmLRaGhodqwYcNd9edu20xNTVVycrLNAgAAAADZcU/hKoOjo6NatGihxYsX3/E2Z8+eVVpamjw9PW3KPT09lZCQcFf9uNs2o6OjVbBgQevi6+t7V/sHAAAAkHuZEq5yugEDBigpKcm6HDt2zN5dAgAAAJDDZOtLhM3k4eEhR0dHJSYm2pQnJibedLKK+9Wms7PzTZ/hAgAAAIA7YbcrV05OTqpSpYpWr15tLUtPT9fq1atVo0aNh6ZNAAAAALgTdrtyJUmRkZEKDw9X1apVVb16dcXGxiolJUURERGSpE6dOqlEiRKKjo6W9PeEFbt377b+fOLECW3dulX58uVT6dKl76hNAAAAALgf7Bqu2rVrpzNnzmjw4MFKSEhQcHCwli9fbp2Q4ujRo7JY/rm4dvLkSVWqVMn6euzYsRo7dqzq1KmjtWvX3lGbAAAAAHA/2DVcSVLv3r3Vu3fvLNdlBKYM/v7+MgzjntoEAAAAgPuB2QIBAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABMQrgAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAEhCsAAAAAMAHhCgAAAABMQLgCAAAAABM8FOFq0qRJ8vf3l4uLi0JCQrRp06Zb1p83b57KlSsnFxcXBQYGaunSpTbrO3fuLAcHB5ulYcOG93MIAAAAAHI5u4eruLg4RUZGKioqSvHx8QoKClJYWJhOnz6dZf2ff/5ZHTp0UJcuXbRlyxa1aNFCLVq00M6dO23qNWzYUKdOnbIuc+bMeRDDAQAAAJBL2T1cxcTEqFu3boqIiFCFChU0ZcoUubm5adq0aVnW//DDD9WwYUP1799f5cuX1/Dhw1W5cmVNnDjRpp6zs7O8vLysS+HChR/EcAAAAADkUnYNV1evXtXmzZsVGhpqLbNYLAoNDdWGDRuy3GbDhg029SUpLCwsU/21a9eqePHiCggI0Kuvvqpz586ZPwAAAAAA+H957Lnzs2fPKi0tTZ6enjblnp6e2rt3b5bbJCQkZFk/ISHB+rphw4Z6/vnnVbJkSR08eFADBw5Uo0aNtGHDBjk6OmZqMzU1VampqdbXycnJ9zIsAAAAALmQXcPV/dK+fXvrz4GBgapYsaIee+wxrV27Vs8++2ym+tHR0Ro6dOiD7CIAAACAfxm73hbo4eEhR0dHJSYm2pQnJibKy8sry228vLyyVV+SSpUqJQ8PDx04cCDL9QMGDFBSUpJ1OXbsWDZHAgAAACC3s2u4cnJyUpUqVbR69WprWXp6ulavXq0aNWpkuU2NGjVs6kvSypUrb1pfko4fP65z587J29s7y/XOzs4qUKCAzQIAAAAA2WH32QIjIyM1depUzZw5U3v27NGrr76qlJQURURESJI6deqkAQMGWOv37dtXy5cv17hx47R3714NGTJEv/32m3r37i1JunTpkvr3769ffvlFR44c0erVq9W8eXOVLl1aYWFhdhkjAAAAgH8/uz9z1a5dO505c0aDBw9WQkKCgoODtXz5cuukFUePHpXF8k8GfOqppzR79mwNGjRIAwcOVJkyZbRw4UI98cQTkiRHR0dt375dM2fO1IULF+Tj46MGDRpo+PDhcnZ2tssYAQAAAPz72T1cSVLv3r2tV55utHbt2kxlbdq0UZs2bbKs7+rqqu+++87M7gEAAADAbdn9tkAAAAAA+DcgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJiBcAQAAAIAJCFcAAAAAYALCFQAAAACYgHAFAAAAACYgXAEAAACACQhXAAAAAGACwhUAAAAAmIBwBQAAAAAmIFwBAAAAgAkIVwAAAABgAsIVAAAAAJiAcAUAAAAAJngowtWkSZPk7+8vFxcXhYSEaNOmTbesP2/ePJUrV04uLi4KDAzU0qVLbdYbhqHBgwfL29tbrq6uCg0N1f79++/nEAAAAADkcnYPV3FxcYqMjFRUVJTi4+MVFBSksLAwnT59Osv6P//8szp06KAuXbpoy5YtatGihVq0aKGdO3da64wePVoTJkzQlClTtHHjRrm7uyssLExXrlx5UMMCAAAAkMvYPVzFxMSoW7duioiIUIUKFTRlyhS5ublp2rRpWdb/8MMP1bBhQ/Xv31/ly5fX8OHDVblyZU2cOFHS31etYmNjNWjQIDVv3lwVK1bUF198oZMnT2rhwoUPcGQAAAAAcpM89tz51atXtXnzZg0YMMBaZrFYFBoaqg0bNmS5zYYNGxQZGWlTFhYWZg1Ohw8fVkJCgkJDQ63rCxYsqJCQEG3YsEHt27fP1GZqaqpSU1Otr5OSkiRJycnJdz02M6Wn/mnvLgAPvYflfL1XnO/Arf1bznWJ8x24nYflfM/oh2EYt61r13B19uxZpaWlydPT06bc09NTe/fuzXKbhISELOsnJCRY12eU3azOjaKjozV06NBM5b6+vnc2EAB2VzDW3j0A8CBwrgO5x8N2vl+8eFEFCxa8ZR27hquHxYABA2yuhqWnp+v8+fMqWrSoHBwc7NgzPIySk5Pl6+urY8eOqUCBAvbuDoD7iPMdyB0413ErhmHo4sWL8vHxuW1du4YrDw8POTo6KjEx0aY8MTFRXl5eWW7j5eV1y/oZfyYmJsrb29umTnBwcJZtOjs7y9nZ2aasUKFC2RkKcqECBQrwBgzkEpzvQO7AuY6bud0Vqwx2ndDCyclJVapU0erVq61l6enpWr16tWrUqJHlNjVq1LCpL0krV6601i9ZsqS8vLxs6iQnJ2vjxo03bRMAAAAA7pXdbwuMjIxUeHi4qlatqurVqys2NlYpKSmKiIiQJHXq1EklSpRQdHS0JKlv376qU6eOxo0bpyZNmmju3Ln67bff9Omnn0qSHBwc1K9fP40YMUJlypRRyZIl9d5778nHx0ctWrSw1zABAAAA/MvZPVy1a9dOZ86c0eDBg5WQkKDg4GAtX77cOiHF0aNHZbH8c4Htqaee0uzZszVo0CANHDhQZcqU0cKFC/XEE09Y67z11ltKSUlR9+7ddeHCBdWqVUvLly+Xi4vLAx8f/n2cnZ0VFRWV6VZSAP8+nO9A7sC5DrM4GHcypyAAAAAA4Jbs/iXCAAAAAPBvQLgCAAAAABMQrgAAAADABIQrPBSOHDkiBwcHbd269aZ11q5dKwcHB124cOGB9etW/P39FRsbe9/3k9WxWb9+vQIDA5U3b161aNHioTs2wJ261/NoxowZfC/hTTyo9yjAnhwcHLRw4UJ7dwOwIlzhvuvcubMcHBzk4OCgvHnzqmTJknrrrbd05coVax1fX1+dOnXKZtZHe0pOTta7776rcuXKycXFRV5eXgoNDdU333yjBz0HTFbHJjIyUsHBwTp8+LBmzJihp556SqdOnbrjL7gD7kTnzp3v+1dY/Prrr+revfsd1c0qLLRr106///77Xe9/xowZ1vcni8Uib29vtWvXTkePHr3rNh8W2Tm2wN26k8/4f6Prx339cuDAAbv2ia8dsj+7T8WO3KFhw4aaPn26rl27ps2bNys8PFwODg764IMPJEmOjo7y8vKycy//ljF9f1JSkkaMGKFq1aopT548+uGHH/TWW2+pXr16D/Q35Vkdm4MHD6pHjx565JFHrGX3evyuXr0qJyene2oDyK5ixYrd0/aurq5ydXW9pzYKFCigffv2yTAMHT58WD179lSbNm20cePGe2r3dq5du6a8efPet/bv9dgCd+p2n/H/Vhnjvt7dnnd8Bv97cOUKD4Szs7O8vLzk6+urFi1aKDQ0VCtXrrSuz+rWt6VLl6ps2bJydXXVM888oyNHjmRqd+rUqfL19ZWbm5tatmypmJiYTMFn0aJFqly5slxcXFSqVCkNHTpUf/311037OnDgQB05ckQbN25UeHi4KlSooLJly6pbt27aunWr8uXLl+V2MTExCgwMlLu7u3x9fdWzZ09dunTJuv5///ufmjVrpsKFC8vd3V2PP/64li5dKkn6448/1LFjRxUrVkyurq4qU6aM9Q37+mOT8fO5c+f08ssvy8HBQTNmzMjytsCffvpJtWvXlqurq3x9fdWnTx+lpKRY1/v7+2v48OHq1KmTChQowG+4kW0//PCDqlevLmdnZ3l7e+udd96xObcuXryojh07yt3dXd7e3ho/frzq1q2rfv36WetcfzXKMAwNGTJEjz76qJydneXj46M+ffpIkurWrav//e9/ev31162/IZayvi3wv//9r6pVqyYXFxd5eHioZcuWtxyHg4ODvLy85O3traeeekpdunTRpk2blJycbK1zu/eRvXv3qlatWnJxcVGFChW0atUqm9uVMs7duLg41alTRy4uLpo1a5Yk6bPPPlP58uXl4uKicuXK6eOPP7a2e/XqVfXu3Vve3t5ycXGRn5+foqOjb3u8bjy20t/fG9m8eXPly5dPBQoUUNu2bZWYmGhdP2TIEAUHB+vLL7+Uv7+/ChYsqPbt2+vixYu3PH7A7T7jz507pw4dOqhEiRJyc3NTYGCg5syZY9NG3bp11adPH7311lsqUqSIvLy8NGTIEJs6+/fv19NPP209z67fR4YdO3aoXr16cnV1VdGiRdW9e3ebz+KMqzsjR46Up6enChUqpGHDhumvv/5S//79VaRIET3yyCOZQtOtxn394ujoKOn2749169ZV79691a9fP3l4eCgsLEyStHPnTjVq1Ej58uWTp6enXnrpJZ09e9a63fz58xUYGGgdX2hoqFJSUjRkyBDNnDlTixYtsr5Hrl279rZjgPkIV3jgdu7cqZ9//vmWv6E5duyYnn/+eTVr1kxbt25V165d9c4779jUWb9+vXr06KG+fftq69atql+/vt5//32bOj/++KM6deqkvn37avfu3frkk080Y8aMTPUypKena+7cuerYsaN8fHwyrc+XL5/y5Mn6gq/FYtGECRO0a9cuzZw5U99//73eeust6/pevXopNTVV69at044dO/TBBx9Yg9p7772n3bt3a9myZdqzZ48mT54sDw+PTPvIuEWwQIECio2N1alTp9SuXbtM9Q4ePKiGDRuqVatW2r59u+Li4vTTTz+pd+/eNvXGjh2roKAgbdmyRe+9916W4wKycuLECTVu3FjVqlXTtm3bNHnyZH3++ecaMWKEtU5kZKTWr1+vxYsXa+XKlfrxxx8VHx9/0za//vprjR8/Xp988on279+vhQsXKjAwUJL0zTff6JFHHtGwYcN06tQpnTp1Kss2lixZopYtW6px48basmWLVq9ererVq9/xuE6fPq0FCxbI0dHR+p+k272PpKWlqUWLFnJzc9PGjRv16aef6t13382y/XfeeUd9+/bVnj17FBYWplmzZmnw4MF6//33tWfPHo0cOVLvvfeeZs6cKUmaMGGCFi9erP/85z/at2+fZs2aJX9//9serxulp6erefPmOn/+vH744QetXLlShw4dyvT+cfDgQS1cuFDffvutvv32W/3www8aNWrUHR8/IKvP+CtXrqhKlSpasmSJdu7cqe7du+ull17Spk2bbLadOXOm3N3dtXHjRo0ePVrDhg2zBqj09HQ9//zzcnJy0saNGzVlyhS9/fbbNtunpKQoLCxMhQsX1q+//qp58+Zp1apVmT77vv/+e508eVLr1q1TTEyMoqKi1LRpUxUuXFgbN25Ujx499Morr+j48eN3dQzu5P0xY7xOTk5av369pkyZogsXLqhevXqqVKmSfvvtNy1fvlyJiYlq27atJOnUqVPq0KGDXn75Ze3Zs0dr167V888/L8Mw9Oabb6pt27Zq2LCh9T3yqaeeuqv+4x4ZwH0WHh5uODo6Gu7u7oazs7MhybBYLMb8+fOtdQ4fPmxIMrZs2WIYhmEMGDDAqFChgk07b7/9tiHJ+OOPPwzDMIx27doZTZo0sanTsWNHo2DBgtbXzz77rDFy5EibOl9++aXh7e2dZV8TExMNSUZMTMxtx+Xn52eMHz/+puvnzZtnFC1a1Po6MDDQGDJkSJZ1mzVrZkRERGS57sZjYxiGUbBgQWP69OnW12vWrLE5Nl26dDG6d+9u086PP/5oWCwW4/Lly9b+t2jR4hYjRG4XHh5uNG/ePMt1AwcONAICAoz09HRr2aRJk4x8+fIZaWlpRnJyspE3b15j3rx51vUXLlww3NzcjL59+1rLrj+Pxo0bZ5QtW9a4evVqlvvM6pybPn26zTlfo0YNo2PHjnc8xunTpxuSDHd3d8PNzc2QZEgy+vTpY61zu/eRZcuWGXny5DFOnTplXb9y5UpDkrFgwQLDMP45j2NjY23aeeyxx4zZs2fblA0fPtyoUaOGYRiG8dprrxn16tWzOc4ZsnO8VqxYYTg6OhpHjx61rt+1a5chydi0aZNhGIYRFRVluLm5GcnJydY6/fv3N0JCQrJsHzCMO/uMz0qTJk2MN954w/q6Tp06Rq1atWzqVKtWzXj77bcNwzCM7777zsiTJ49x4sQJ6/ply5bZnGeffvqpUbhwYePSpUvWOkuWLDEsFouRkJBg7a+fn5+RlpZmrRMQEGDUrl3b+vqvv/4y3N3djTlz5tzRuDOW1q1bG4Zx+/fHjPFWqlTJps3hw4cbDRo0sCk7duyYIcnYt2+fsXnzZkOSceTIkZv26Wbv2XhweOYKD8QzzzyjyZMnKyUlRePHj1eePHnUqlWrm9bfs2ePQkJCbMpq1Khh83rfvn2ZbvepXr26vv32W+vrbdu2af369TZXqtLS0nTlyhX9+eefcnNzs9neuIfJKlatWqXo6Gjt3btXycnJ+uuvv2z206dPH7366qtasWKFQkND1apVK1WsWFGS9Oqrr6pVq1aKj49XgwYN1KJFi3v6jdO2bdu0fft2621HGWNLT0/X4cOHVb58eUlS1apV73ofyN327NmjGjVqWG/Pk6SaNWvq0qVLOn78uP744w9du3bN5qpRwYIFFRAQcNM227Rpo9jYWJUqVUoNGzZU48aN1axZs5teLc7K1q1b1a1bt2yNJX/+/IqPj9e1a9e0bNkyzZo1y+Y943bvI/v27ZOvr6/Nc483u1p2/TmXkpKigwcPqkuXLjZ9/uuvv6yT03Tu3Fn169dXQECAGjZsqKZNm6pBgwaSsne89uzZI19fX/n6+lrLKlSooEKFCmnPnj2qVq2apL9vJcyfP7+1jre3t06fPn1nBxK51u0+49PS0jRy5Ej95z//0YkTJ3T16lWlpqZm+gzO+EzMcP2/v4x/w9ffVXLj/wv27NmjoKAgubu7W8tq1qyp9PR07du3T56enpKkxx9/XBbLPzdveXp62kwa5ejoqKJFi972337GuDNk7Pd274+PPvqoJKlKlSo27W3btk1r1qzJ8vGDgwcPqkGDBnr22WcVGBiosLAwNWjQQK1bt1bhwoVv2U88WNwWiAfC3d1dpUuXVlBQkKZNm6aNGzfq888/v+/7vXTpkoYOHaqtW7dalx07dmj//v1ycXHJVL9YsWIqVKiQ9u7dm639HDlyRE2bNlXFihX19ddfa/PmzZo0aZKkv5+ZkKSuXbvq0KFDeumll7Rjxw5VrVpVH330kSSpUaNG1udJTp48qWeffVZvvvnmPY37lVdesRn3tm3btH//fj322GPWetd/AAH25uvrq3379unjjz+Wq6urevbsqaefflrXrl274zbuZnILi8Wi0qVLq3z58oqMjNSTTz6pV1991bo+u+8jt3L9OZfxHMjUqVNt2t65c6d++eUXSVLlypV1+PBhDR8+XJcvX1bbtm3VunVrSeYcrxvdOMGGg4OD0tPT77o95A63+4wfM2aMPvzwQ7399ttas2aNtm7dqrCwMOvnY4YH9e8vq/3czb4zxp2xeHt7Z6sfN34GX7p0yfo4xPVLxrNmjo6OWrlypZYtW6YKFSroo48+UkBAgA4fPpyt/eL+IlzhgbNYLBo4cKAGDRqky5cvZ1mnfPnyme7FzvjPRoaAgAD9+uuvNmU3vq5cubL27dtn8+aXsVz/W6vr+9a+fXvNmjVLJ0+ezLT+0qVLWU6GsXnzZqWnp2vcuHF68sknVbZs2Sy39/X1VY8ePfTNN9/ojTfe0NSpU63rihUrpvDwcH311VeKjY3Vp59+msWRuTOVK1fW7t27sxw3sxHBDOXLl9eGDRtsrvauX79e+fPn1yOPPKJSpUopb968NudkUlLSbadNd3V1VbNmzTRhwgStXbtWGzZs0I4dOyRJTk5OSktLu+X2FStW1OrVq+9hZH8/FxUXF2d9Pux27yMBAQE6duyYzeQQN74XZcXT01M+Pj46dOhQpnZLlixprVegQAG1a9dOU6dOVVxcnL7++mudP39e0q2P1/XKly+vY8eO6dixY9ay3bt368KFC6pQocJdHyvgRll9xq9fv17NmzfXiy++qKCgIJUqVSrbX6GQ8W/4+uctb/x/Qfny5bVt2zabyZvWr19vPU8flNu9P95M5cqVtWvXLvn7+2d6T8gIYg4ODqpZs6aGDh2qLVu2yMnJSQsWLJB0Z++RuP8IV7CLNm3ayNHR0Xp150Y9evTQ/v371b9/f+3bt0+zZ8/WjBkzbOq89tprWrp0qWJiYrR//3598sknWrZsmc1l+MGDB+uLL77Q0KFDtWvXLu3Zs0dz587VoEGDbtq3999/X76+vgoJCdEXX3yh3bt3a//+/Zo2bZoqVapkM+tQhtKlS+vatWv66KOPdOjQIX355ZeaMmWKTZ1+/frpu+++0+HDhxUfH681a9ZYb88bPHiwFi1apAMHDmjXrl369ttvrevuxttvv62ff/5ZvXv3tv7Wa9GiRZke6gVuJykpKdNvUY8dO6aePXvq2LFjeu2117R3714tWrRIUVFRioyMlMViUf78+RUeHq7+/ftrzZo12rVrl7p06SKLxWJzjl5vxowZ+vzzz7Vz504dOnRIX331lVxdXeXn5yfp71vW1q1bpxMnTtjMnnW9qKgozZkzR1FRUdqzZ4918pjs8PX1VcuWLTV48GBJt38fqV+/vh577DGFh4dr+/btWr9+vXXdzcaaYejQoYqOjtaECRP0+++/a8eOHZo+fbpiYmIk/T0L6Zw5c7R37179/vvvmjdvnry8vFSoUKHbHq/rhYaGKjAwUB07dlR8fLw2bdqkTp06qU6dOtweDNPd+BlfpkwZrVy5Uj///LP27NmjV155xeaXEXciNDRUZcuWVXh4uLZt26Yff/wx08QxHTt2lIuLi8LDw7Vz506tWbNGr732ml566SXrLYEPwu3eH2+mV69eOn/+vDp06KBff/1VBw8e1HfffaeIiAilpaVp48aNGjlypH777TcdPXpU33zzjc6cOWP9/4K/v7+2b9+uffv26ezZs/d0FRt3j3AFu8iTJ4969+6t0aNH2/yGKcOjjz6qr7/+WgsXLlRQUJCmTJmikSNH2tSpWbOmpkyZopiYGAUFBWn58uV6/fXXbW7TCQsL07fffqsVK1aoWrVqevLJJzV+/Pgs//ORoUiRIvrll1/04osvasSIEapUqZJq166tOXPmaMyYMVl+UW9QUJBiYmL0wQcf6IknntCsWbOs0yVnSEtLU69evVS+fHk1bNhQZcuWtU657OTkpAEDBqhixYrWS/9z587N1jG9XsWKFfXDDz/o999/V+3atVWpUiUNHjw4yxkQgVtZu3atKlWqZLMMHTpUJUqU0NKlS7Vp0yYFBQWpR48e6tKli80vLmJiYlSjRg01bdpUoaGhqlmzpnXK8awUKlRIU6dOVc2aNVWxYkWtWrVK//3vf1W0aFFJ0rBhw3TkyBE99thjN/0umbp162revHlavHixgoODVa9evUxXwe/E66+/riVLlmjTpk23fR9xdHTUwoULdenSJVWrVk1du3a1/qfvdrcNdu3aVZ999pmmT5+uwMBA1alTRzNmzLBeucqfP79Gjx6tqlWrqlq1ajpy5IiWLl0qi8Vy2+N1PQcHBy1atEiFCxfW008/rdDQUJUqVUpxcXHZPjbA7dz4GT9o0CBVrlxZYWFhqlu3rry8vLL9ZbcWi0ULFizQ5cuXVb16dXXt2jXTzL9ubm767rvvdP78eVWrVk2tW7fWs88+q4kTJ5o4utu7k/fHrPj4+Gj9+vVKS0tTgwYNFBgYqH79+qlQoUKyWCwqUKCA1q1bp8aNG6ts2bIaNGiQxo0bp0aNGkmSunXrpoCAAFWtWlXFihXT+vXrH8RwcQMH416e4AceMt26ddPevXv1448/2rsrAG6QkpKiEiVKaNy4cerSpYu9u3NfrV+/XrVq1dKBAwdsnnMEAPy7MVsgcrSxY8eqfv36cnd317JlyzRz5kybL+AEYD9btmzR3r17Vb16dSUlJWnYsGGSpObNm9u5Z+ZbsGCB8uXLpzJlyujAgQPq27evatasSbACgFyGcIUcbdOmTRo9erQuXryoUqVKacKECeratau9uwXg/40dO1b79u2Tk5OTqlSpoh9//DHLL8jO6S5evKi3335bR48elYeHh0JDQzVu3Dh7dwsA8IBxWyAAAAAAmIAJLQAAAADABIQrAAAAADAB4QoAAAAATEC4AgAAAAATEK4AAAAAwASEKwAAAAAwAeEKAAAAAExAuAIAAAAAExCuAAAAAMAE/wdoL9GQjjumqAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# simple graph to see the comparision of the models\n",
    "# Ridge Classifier\n",
    "rc = RidgeClassifier(class_weight='balanced')\n",
    "classifier = MultiOutputClassifier(rc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "rc_results = model_eval(y_test, classifier_preds, emotion).iloc[-1]\n",
    "\n",
    "# Logistic Regression\n",
    "lc = LogisticRegression(max_iter=1000)\n",
    "classifier = MultiOutputClassifier(lc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "lc_results = model_eval(y_test, classifier_preds, emotion).iloc[-1]\n",
    "\n",
    "# Random Forest\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "classifier = MultiOutputClassifier(rf, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "rf_results = model_eval(y_test, classifier_preds, emotion).iloc[-1]\n",
    "\n",
    "# Plotting the results\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.plot(rc_results, label=\"Ridge Classifier\")\n",
    "plt.plot(lc_results, label=\"Logistic Regression\")\n",
    "plt.plot(rf_results, label=\"Random Forest\")\n",
    "plt.xticks(rotation=45)\n",
    "plt.title(\"Model Comparison\")\n",
    "plt.ylabel(\"F1 Score\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# also use the model accuracy to compare the models\n",
    "# Ridge Classifier\n",
    "rc = RidgeClassifier(class_weight='balanced')\n",
    "classifier = MultiOutputClassifier(rc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "rc_accuracy = accuracy_score(y_test, classifier_preds)\n",
    "\n",
    "# Logistic Regression\n",
    "lc = LogisticRegression(max_iter=1000)\n",
    "classifier = MultiOutputClassifier(lc, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "lc_accuracy = accuracy_score(y_test, classifier_preds)\n",
    "\n",
    "# Random Forest\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "classifier = MultiOutputClassifier(rf, n_jobs=-1)\n",
    "classifier.fit(X_train, y_train)\n",
    "classifier_preds = classifier.predict(X_test)\n",
    "rf_accuracy = accuracy_score(y_test, classifier_preds)\n",
    "\n",
    "# Plotting the results\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.bar([\"Ridge Classifier\", \"Logistic Regression\", \"Random Forest\"], [rc_accuracy, lc_accuracy, rf_accuracy])\n",
    "plt.title(\"Model Comparison\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ubDuS41V5-xc"
   },
   "source": [
    "Making predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "QEj3RilI46Wy"
   },
   "outputs": [],
   "source": [
    "# Retrieving initial text preprocessings\n",
    "def preprocess_data(x):\n",
    "\n",
    "    # Adding a space between words and punctation\n",
    "    x = re.sub( r'([a-zA-Z\\[\\]])([,;.!?])', r'\\1 \\2', x)\n",
    "    x = re.sub( r'([,;.!?])([a-zA-Z\\[\\]])', r'\\1 \\2', x)\n",
    "\n",
    "    # Demojize\n",
    "    x = emoji.demojize(x)\n",
    "\n",
    "    # Expand contraction\n",
    "    x = contractions.fix(x)\n",
    "\n",
    "    # Lower\n",
    "    x = x.lower()\n",
    "\n",
    "    #correct some acronyms/typos/abbreviations\n",
    "    x = re.sub(r\"lmao\", \"laughing my ass off\", x)\n",
    "    x = re.sub(r\"amirite\", \"am i right\", x)\n",
    "    x = re.sub(r\"\\b(tho)\\b\", \"though\", x)\n",
    "    x = re.sub(r\"\\b(ikr)\\b\", \"i know right\", x)\n",
    "    x = re.sub(r\"\\b(ya|u)\\b\", \"you\", x)\n",
    "    x = re.sub(r\"\\b(eu)\\b\", \"europe\", x)\n",
    "    x = re.sub(r\"\\b(da)\\b\", \"the\", x)\n",
    "    x = re.sub(r\"\\b(dat)\\b\", \"that\", x)\n",
    "    x = re.sub(r\"\\b(dats)\\b\", \"that is\", x)\n",
    "    x = re.sub(r\"\\b(cuz)\\b\", \"because\", x)\n",
    "    x = re.sub(r\"\\b(fkn)\\b\", \"fucking\", x)\n",
    "    x = re.sub(r\"\\b(tbh)\\b\", \"to be honest\", x)\n",
    "    x = re.sub(r\"\\b(tbf)\\b\", \"to be fair\", x)\n",
    "    x = re.sub(r\"faux pas\", \"mistake\", x)\n",
    "    x = re.sub(r\"\\b(btw)\\b\", \"by the way\", x)\n",
    "    x = re.sub(r\"\\b(bs)\\b\", \"bullshit\", x)\n",
    "    x = re.sub(r\"\\b(kinda)\\b\", \"kind of\", x)\n",
    "    x = re.sub(r\"\\b(bruh)\\b\", \"bro\", x)\n",
    "    x = re.sub(r\"\\b(w/e)\\b\", \"whatever\", x)\n",
    "    x = re.sub(r\"\\b(w/)\\b\", \"with\", x)\n",
    "    x = re.sub(r\"\\b(w/o)\\b\", \"without\", x)\n",
    "    x = re.sub(r\"\\b(doj)\\b\", \"department of justice\", x)\n",
    "\n",
    "    # replace some words with multiple occurences of a letter, example \"coooool\" turns into --> cool\n",
    "    x = re.sub(r\"\\b(j+e{2,}z+e*)\\b\", \"jeez\", x)\n",
    "    x = re.sub(r\"\\b(co+l+)\\b\", \"cool\", x)\n",
    "    x = re.sub(r\"\\b(g+o+a+l+)\\b\", \"goal\", x)\n",
    "    x = re.sub(r\"\\b(s+h+i+t+)\\b\", \"shit\", x)\n",
    "    x = re.sub(r\"\\b(o+m+g+)\\b\", \"omg\", x)\n",
    "    x = re.sub(r\"\\b(w+t+f+)\\b\", \"wtf\", x)\n",
    "    x = re.sub(r\"\\b(w+h+a+t+)\\b\", \"what\", x)\n",
    "    x = re.sub(r\"\\b(y+e+y+|y+a+y+|y+e+a+h+)\\b\", \"yeah\", x)\n",
    "    x = re.sub(r\"\\b(w+o+w+)\\b\", \"wow\", x)\n",
    "    x = re.sub(r\"\\b(w+h+y+)\\b\", \"why\", x)\n",
    "    x = re.sub(r\"\\b(s+o+)\\b\", \"so\", x)\n",
    "    x = re.sub(r\"\\b(f)\\b\", \"fuck\", x)\n",
    "    x = re.sub(r\"\\b(w+h+o+p+s+)\\b\", \"whoops\", x)\n",
    "    x = re.sub(r\"\\b(ofc)\\b\", \"of course\", x)\n",
    "    x = re.sub(r\"\\b(the us)\\b\", \"usa\", x)\n",
    "    x = re.sub(r\"\\b(gf)\\b\", \"girlfriend\", x)\n",
    "    x = re.sub(r\"\\b(hr)\\b\", \"human ressources\", x)\n",
    "    x = re.sub(r\"\\b(mh)\\b\", \"mental health\", x)\n",
    "    x = re.sub(r\"\\b(idk)\\b\", \"i do not know\", x)\n",
    "    x = re.sub(r\"\\b(gotcha)\\b\", \"i got you\", x)\n",
    "    x = re.sub(r\"\\b(y+e+p+)\\b\", \"yes\", x)\n",
    "    x = re.sub(r\"\\b(a*ha+h[ha]*|a*ha +h[ha]*)\\b\", \"haha\", x)\n",
    "    x = re.sub(r\"\\b(o?l+o+l+[ol]*)\\b\", \"lol\", x)\n",
    "    x = re.sub(r\"\\b(o*ho+h[ho]*|o*ho +h[ho]*)\\b\", \"ohoh\", x)\n",
    "    x = re.sub(r\"\\b(o+h+)\\b\", \"oh\", x)\n",
    "    x = re.sub(r\"\\b(a+h+)\\b\", \"ah\", x)\n",
    "    x = re.sub(r\"\\b(u+h+)\\b\", \"uh\", x)\n",
    "\n",
    "    # Handling emojis\n",
    "    x = re.sub(r\"<3\", \" love \", x)\n",
    "    x = re.sub(r\"xd\", \" smiling_face_with_open_mouth_and_tightly_closed_eyes \", x)\n",
    "    x = re.sub(r\":\\)\", \" smiling_face \", x)\n",
    "    x = re.sub(r\"^_^\", \" smiling_face \", x)\n",
    "    x = re.sub(r\"\\*_\\*\", \" star_struck \", x)\n",
    "    x = re.sub(r\":\\(\", \" frowning_face \", x)\n",
    "    x = re.sub(r\":\\^\\(\", \" frowning_face \", x)\n",
    "    x = re.sub(r\";\\(\", \" frowning_face \", x)\n",
    "    x = re.sub(r\":\\/\",  \" confused_face\", x)\n",
    "    x = re.sub(r\";\\)\",  \" wink\", x)\n",
    "    x = re.sub(r\">__<\",  \" unamused \", x)\n",
    "    x = re.sub(r\"\\b([xo]+x*)\\b\", \" xoxo \", x)\n",
    "    x = re.sub(r\"\\b(n+a+h+)\\b\", \"no\", x)\n",
    "\n",
    "    # Handling special cases of text\n",
    "    x = re.sub(r\"h a m b e r d e r s\", \"hamberders\", x)\n",
    "    x = re.sub(r\"b e n\", \"ben\", x)\n",
    "    x = re.sub(r\"s a t i r e\", \"satire\", x)\n",
    "    x = re.sub(r\"y i k e s\", \"yikes\", x)\n",
    "    x = re.sub(r\"s p o i l e r\", \"spoiler\", x)\n",
    "    x = re.sub(r\"thankyou\", \"thank you\", x)\n",
    "    x = re.sub(r\"a^r^o^o^o^o^o^o^o^n^d\", \"around\", x)\n",
    "\n",
    "    # Remove special characters and numbers replace by space + remove double space\n",
    "    x = re.sub(r\"\\b([.]{3,})\",\" dots \", x)\n",
    "    x = re.sub(r\"[^A-Za-z!?_]+\",\" \", x)\n",
    "    x = re.sub(r\"\\b([s])\\b *\",\"\", x)\n",
    "    x = re.sub(r\" +\",\" \", x)\n",
    "    x = x.strip()\n",
    "\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "3VAB3U_K46Wz"
   },
   "outputs": [],
   "source": [
    "def predict_samples(text_samples, model):\n",
    "\n",
    "    # Text preprocessing and cleaning\n",
    "    text_samples = pd.Series(text_samples)\n",
    "    text_samples_clean = text_samples.apply(preprocess_data)\n",
    "\n",
    "    # Create tfidf representation\n",
    "    tfidf_text_samples_clean = vectorizer.transform(text_samples_clean)\n",
    "\n",
    "    # labels predictions\n",
    "    samples_pred_labels = model.predict(tfidf_text_samples_clean)\n",
    "    samples_pred_labels_df = pd.DataFrame(samples_pred_labels)\n",
    "    samples_pred_labels_df = samples_pred_labels_df.apply(lambda x: [emotion[i] for i in range(len(x)) if x[i]==1], axis=1)\n",
    "\n",
    "    return pd.DataFrame({\"Text\":text_samples, \"Emotions\":list(samples_pred_labels_df)})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WBnRiDhf6DP7"
   },
   "source": [
    "Evaluation and Prediction testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "NKVH65a-46W0",
    "outputId": "23add23d-7fb7-4fa3-afde-defd99fd2582"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Text</th>\n",
       "      <th>Emotions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Was it about my model?</td>\n",
       "      <td>[fear]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Text Emotions\n",
       "0  Was it about my model?   [fear]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict samples\n",
    "predict_samples(\"Was it about my model?\", classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Text</th>\n",
       "      <th>Emotions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>I am so happy today</td>\n",
       "      <td>[joy, sadness]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>I am feeling very sad</td>\n",
       "      <td>[fear, sadness]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>I am feeling very angry</td>\n",
       "      <td>[anger, fear]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>I am feeling very surprised</td>\n",
       "      <td>[fear]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>I am feeling very fearful</td>\n",
       "      <td>[fear]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Text         Emotions\n",
       "0          I am so happy today   [joy, sadness]\n",
       "1        I am feeling very sad  [fear, sadness]\n",
       "2      I am feeling very angry    [anger, fear]\n",
       "3  I am feeling very surprised           [fear]\n",
       "4    I am feeling very fearful           [fear]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict samples for paragraph\n",
    "text_samples = [\"I am so happy today\", \"I am feeling very sad\", \"I am feeling very angry\", \"I am feeling very surprised\", \"I am feeling very fearful\"]\n",
    "predict_samples(text_samples, classifier)\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}