glove.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "77013426-68e6-4585-9c35-a06128fcb0c2",
   "metadata": {},
   "source": [
    "## **Sentiment Analysis** of the **Restaurant Reviews** from **YELP** dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3253ef45-46d6-4e25-9068-65cf80785b2f",
   "metadata": {},
   "source": [
    "###  An example of a **pretrained word embeddings**: **Glove**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93226080-1c2c-4ce7-a5c6-57ec6efa88ea",
   "metadata": {},
   "source": [
    "#### Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9623e81a-f3e5-47b3-8fc1-49b6252b0092",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "from zeugma.embeddings import EmbeddingTransformer\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "from sklearn.metrics import accuracy_score,f1_score,precision_score, recall_score, confusion_matrix\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn import svm\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import sys\n",
    "import os\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b348df5d-6d44-4865-b369-b7562f73e3ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "if not sys.warnoptions:\n",
    "    warnings.simplefilter(\"ignore\")\n",
    "    os.environ[\"PYTHONWARNINGS\"] = \"ignore\" "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b86668fb-0828-405a-bd1e-52cb7d2d652f",
   "metadata": {},
   "source": [
    "#### Read the dataset from a **GCP bucket**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7d8b0835-230b-4f54-a956-2845bccb1790",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('gs://demo2222/df_final.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c870344e-d6d2-43dc-8bfe-0cafa9037606",
   "metadata": {},
   "source": [
    "#### Train/test split **keeping** the two **columns of interest** and **preserve** the **same proportions** of examples in **each class** as **observed** in the **original dataset**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ccca595f-3bce-4284-9276-62e3ada80567",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df['cleaned_reviews'] \n",
    "y = df['sentiment']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 42, stratify=y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83b515ff-6d79-414b-8f63-2a84ec25589d",
   "metadata": {},
   "source": [
    "#### UDF for **evaluating** models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "88f36032-ea20-4678-aa5a-5cee21c92e69",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate(y_test, predictions):\n",
    "    cf_matrix = confusion_matrix(y_test, predictions)\n",
    "    sns.heatmap(cf_matrix, annot = True, fmt = 'd',cmap=\"Blues\")\n",
    "    plt.title('Heatmap of confusion matrix for Test data')\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fbf4bb88-54bd-4cc9-8fb2-547b10f05f6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus_train = X_train\n",
    "corpus_test = X_test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5a3f73b-b810-4f1c-9c7a-0809a4f251e4",
   "metadata": {},
   "source": [
    "### **glove-twitter-200**: pre-trained **glove vectors** based on 2B tweets, 27B tokens, 1.2M vocab, **uncased**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cec7cbd4-f673-4afa-a012-09985c663d0f",
   "metadata": {},
   "outputs": [],
   "source": [
    "glove_twitter = EmbeddingTransformer('glove-twitter-200') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "24530bef-ee5f-4393-9d46-0d798896b897",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train_twitter = glove_twitter.transform(corpus_train) \n",
    "x_test_twitter = glove_twitter.transform(corpus_test) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "753a7793-e10e-4d01-973b-832eaa98eb2e",
   "metadata": {},
   "source": [
    "#### Using **Logistic Regression** model with the **default hyperparameters**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8e6012cb-587a-4636-8745-fb6f209f00fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_lr = LogisticRegression(random_state=42)\n",
    "model_lr.fit(x_train_twitter, y_train)\n",
    "predictions = model_lr.predict(x_test_twitter)\n",
    "evaluate(y_test, predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "433a08be-31f9-4d19-9535-96f698ff150a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.8993    0.8973    0.8983      1393\n",
      "         1.0     0.8976    0.8995    0.8985      1393\n",
      "\n",
      "    accuracy                         0.8984      2786\n",
      "   macro avg     0.8984    0.8984    0.8984      2786\n",
      "weighted avg     0.8984    0.8984    0.8984      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06a3dc6e-55a1-42db-a46b-1ede0d1a599a",
   "metadata": {},
   "source": [
    "#### Using **SVM** model with the **default hyperparameters**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "abb65571-43db-48d6-a0e9-83b1bf8ae8f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAEWCAYAAACZnQc8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAlrElEQVR4nO3deZwU1b3+8c8zM4AYQEEFEXBJRI1i3FFj9JKIihpFjQvG3IAhP9yNiVHEGDUmqFlv4nVBXCKJUUSjV9w1KHFFRFwQlEjEwCiLCriiLH5/f9QZbIaZoRlm6SmeN696UX1O1Tmnunu+ffrU6SpFBGZmlg9lzd0AMzNrOA7qZmY54qBuZpYjDupmZjnioG5mliMO6mZmOeKg3sJJOkXSPEkfSdqoiersIulxSR9K+v1alHO+pOsbsm3NRdIJkh6u5777SHo9vYZHNHDTmpykmyT9qrnbsa5qkUFd0puS+lZLGyTpyQYqPyRt3RBlNSZJrYA/AAdGRLuIeK+Jqh4CvAt0iIiz61tIRFwaET9suGY1PElbpvdDRV3bRcTfIuLAelZzCXBleg3/r55lACDpgfTh8JGkpZKWFDweUY/yLpZ089q0aTXlj5dU0u+BlqbON6qVvC7AesDUJq53C2Ba+JdrAEiqiIhla1HEFtTzNaxed0QcXJB3E1AZEResRdushWmRPfViSNpM0t8lvSNppqQzC/J6S3pG0iJJcyRdKal1yns8bfZS6t0cJ6mPpEpJ50qan/Y5QtIhkv4laYGk84spP+WHpDMlvSHpXUm/lVTjayGpjaQ/Sno7LX9MadsA09NmiyQ9Wsv+35D0dGrLbEmDUvoGkv6Snp//SLqgqg1V33ok/U7SwvT8HZzybgIGAuem56dv9a/bVc9XweOhkt5KwzXTJe2f0lfqBUo6XNLU1Nbxkr5akPempJ9KelnS+5Juk7ReLcc8SNJTkv4nlfWGpK+n9NnpNRxYsP2hkl6Q9EHKv7iguKr3w6J0vHtXK38BcLEKvimmut6V1CM93im1Y7sa2vpv4MvAPan8Num9Oza9r2ZI+n8F218s6Q5JN0v6ABhU03NQy/PybUkvprY8Lelrdb1GkvoB5wPHpba9VEu5u0ianPa9jayjUZXXUdK96X22MK13T3nDgX2BK1P5V6b0P6XX4QNJz0vat9hjNCAiWtwCvAn0rZY2CHgyrZcBzwMXAq3J/mjeAA5K+bsBe5F9U9kSeBU4q6CsALYueNwHWJbKawX8P+Ad4BagPbAD8Cnw5TUo/zGgE7A58C/gh7Uc6yXABKAzsAnwNPDLlLdlKquiln03Bz4Ejk/t3gjYOeX9Bbg7tX/L1IbBBc/l0nSc5cApwNuAUv5NwK8K6qn+uA9ZDxFgW2A2sFlBm7+S1i8Gbk7r2wAfAwektp4LzABaF7zmE4HN0vP2KnByLcc9KL1eJ6b2/wqYBVwFtAEOTM9Lu4L27kj2vvkaMA84orbnuKD8M9Jr3JaC91/aZjjwaMp7GTi92Pcz8E/garLguDPZe23/gudsKXBEam/bOspd8boAuwLzgT3TczIw1dum2NeoljpaA/8Bfpxet6NT+6rq3Qj4DrA+2XvtduD/CvYfT7X3PvC9tF8FcDYwF1ivueNOS1mavQH1anT2ZvwIWFSwfMIXQX1PYFa1fYYBf66lvLOAuwoe1xTUFwPl6XH7tM2eBds8XxUIiiy/X8HjU4Fxtez7b+CQgscHAW+m9S2pO6gPK6y3IL0c+AzYviDtJGB8Wh8EzCjIWz/Vs2l6vCJY1PK4D18E9a3JgklfoFW1dqwIGMDPgTEFeWXAW0Cfgtf8ewX5vwFG1HLcg4DXCx7vmNrfpSDtPdIHXA37/xH4n9qe41R+9ffXIFYO6q3Se2IK8CDpA7GO93PftN4DWA60L8i/DLip4Dl7vMi/kxWvC3ANqTNQkD8d+K9iX6Na6tiPgg/8lPZ04fuh2vY7AwsLHo+nlg5NwTYLgZ2KOWYv0aKHX46IiA2rFrLAWGULYLP0NXORpEVkXyO7AEjaJn0NnJu+wl4KbLya+t6LiOVpfXH6f15B/mKg3RqUP7tg/T9kPdCabJbyi9m2uh5kHwrVbcwXPazCcrsVPJ5btRIRn6TVdkXWu0JEzCD7ULsYmC9ptKSa2r/ScUbE52TPUY1tIvsQr6s91V8bIqK212tPSY+lIYL3gZNZ/fthdl2ZEbGULKj2An4fKToVYTNgQUR8WJBW/bWps+5abAGcXe1vogdZ77zY16i29r5V7fhWvI6S1pd0rbIhvg/IhrM2lFReW4GSzpb0ahpmWwRswOpfD0taclCvy2xgZmHQj4j2EXFIyr8GeA3oGREdyAK+GrD+YsrvUbC+OVlvpyZvk/1BFrNtdbOBr9SQ/i7ZV+Tq5b5VZLnVfUzWm6+yaWFmRNwSEd9I9QXw6xrKWOk4JYnsOapvm9bELcBYoEdEbACM4IvXq7ZgXGeQltQNuAj4M/B7SW2KbMvbQCdJ7QvSqr829TlBPRsYXu1vYv2IuBXqfI1WV9ccoFt6vQrbW+VssuGdPdPfwn4pvcbnN42fDwWOBTqmDtv7NOzfZ67lNahPBD5IJ3/aSiqX1EvSHim/PfAB8FE6eXVKtf3nkY3D19fqygc4J51E6gH8CLitlrJuBS6QtImkjcnG9YudYvY3oK+kYyVVSNpI0s7pG8cYYLik9pK2AH6yBuVW9yJwiKROkjYl6/UBIGlbSd9KQe1Tsh7y8hrKGAMcmk7QtSILBp+RfZVvbO3JesefSuoNfLcg7x3gc9bg/ZAC3E3ADcBgssD3y2L2jYjZZMd8maT10snMwWSv5dq4Djg5fSuRpC8pO0HcfjWv0TxgS9VyIh94huz8wpnpPXYU0Lsgv30qb5GkTmQfdIWq/621T+W9A1RIuhDoUO+jXgflMqinoHUY2fjdTLKe6fVkX+MAfkr2h/sh2Zu9ekC9GBiVvqYeW48mrK58yE5SPk8WEO8jCwA1+RUwiexk2xRgckpbrYiYBRxCFiAXpLp2StlnkPWw3wCeJOut3lhMuTX4K/AS2djww6x8vG2Ay8leg7lkJ3zPr7Y/ETGd7ATZ/6ZtDwMOi4gl9WzTmjgVuETSh2QfmmMK2vUJ2UnPp9L7Ya8iyjuTbKjv52lY4kTgxDWYxXE82Vj+28BdwEUR8UixB1OTiJhEduL7SrIx6hl8MXOmrtfo9vT/e5Im11DuEuCoVNZC4DjgzoJN/kh2svhdshP+D1Yr4k/A0WlmzBXAQ8ADZCfu/0P2IVOf4aZ1VtVsBmtCkoJsaGZGc7fFzPIllz11M7N1lYO6mVmOePjFzCxH3FM3M8uRkr2gV9tdTvdXCFvFwueubO4mWAlar2Lt57GvScxZ/MKVJTtvvmSDuplZk6p1Kn7L4qBuZgagku18rxEHdTMzcE/dzCxX3FM3M8uRslovHNmiOKibmYGHX8zMciUnwy/5+GgyM1tbKit+WV1R0o3K7oX7SkHabyW9puw+u3dJ2rAgb5iy+9FOl3RQQfpukqakvCuqXbe+Rg7qZmaQ9dSLXVbvJqBftbRHgF4R8TWySwsPy6rV9sAAsnsd9wOuLrgz1DXAEKBnWqqXuQoHdTMzaNCeekQ8TnYPg8K0hyNiWXo4Aeie1vsDoyPis4iYSXat+96SugIdIuKZdF3+v5DdcLxODupmZpDNfilykTRE0qSCZcga1vYDspuBQHb/2cIbgVSmtG5pvXp6nXyi1MwM1mj2S0SMBEbWqxrpZ2S37Ku6RWFN4zlRR3qdHNTNzADKGn/2i6SBwLeB/eOL655XsvKN6LuT3cqwki+GaArT6+ThFzMzaNAx9RqLl/oBQ4HD071vq4wFBkhqI2krshOiEyNiDvChpL3SrJfvk93buE7uqZuZQYPOU5d0K9AH2FhSJXAR2WyXNsAjaWbihIg4OSKmShoDTCMbljktIpanok4hm0nTlmwM/gFWw0HdzAwa9DIBEXF8Dck31LH9cGB4DemTgF5rUreDupkZ+DIBZma5kpPLBDiom5mBe+pmZrninrqZWY64p25mliO+SYaZWY64p25mliMeUzczyxH31M3McsQ9dTOzHHFP3cwsP1TmoG5mlhtF3NO5RXBQNzODmu8z1AI5qJuZ4Z66mVmuOKibmeVImU+UmpnlSD466g7qZmbg4Rczs1xxUDczyxEHdTOzHHFQNzPLEZU5qJuZ5YZ76mZmOeKgbmaWJ/mI6Q7qZmbgnrqZWa44qJuZ5Yiv/WJmlif56KiTj48mM7O1JKnopYiybpQ0X9IrBWmdJD0i6fX0f8eCvGGSZkiaLumggvTdJE1JeVeoiMod1M3MaNigDtwE9KuWdh4wLiJ6AuPSYyRtDwwAdkj7XC2pPO1zDTAE6JmW6mWuwkHdzIyGDeoR8TiwoFpyf2BUWh8FHFGQPjoiPouImcAMoLekrkCHiHgmIgL4S8E+tXJQNzMju0xA0Ys0RNKkgmVIEVV0iYg5AOn/zim9GzC7YLvKlNYtrVdPr5NPlDaTERedwMH79eKdBR+y+zGXAnDpWUdwyH69WLJ0OTMr32XIRTfz/keL2bxrJ1688wL+9Z/5AEyc8iZnDh8NwN1Xnsqmm3Sgorycp174N2dddhuffx7NdlzWcC68YBiP/3M8nTptxJ1337tS3qg/38Affvcbxj/5DB07dgLghuuu5a6/30FZeRlDh13APt/Ytzma3WKtyZTGiBgJjGyoqmuqoo70Ormn3kz+es8E+p921Upp4ya8xm7HXErv4y7j9f/M55wfHLgi743Kd9lrwOXsNeDyFQEd4HtDb2TP4y5nt6OHs0nHdnzngF2b7BiscfU/4iiuufb6VdLnzpnDM08/Tdeum61I+/eMGTx4/33cOfY+rr72ei791S9Yvnx5Uza3xWvgMfWazEtDKqT/56f0SqBHwXbdgbdTevca0uvkoN5Mnpr8bxa8/8lKaeMmvMby5Z8DMHHKTLp12XC15Xz48acAVFSU0aqinGzozfJgt933oMMGG6yS/ttfX8aPzz5npeAy/rFx9DvkUFq3bk337j3o0WMLXpnyclM2t8VrgqA+FhiY1gcCdxekD5DURtJWZCdEJ6Yhmg8l7ZVmvXy/YJ9aNdrwi6TtyE4AdCP7yvA2MDYiXm2sOvPk+/335o6HJ694vGW3jXjm1qF8+PGn/OKqe3nqhX+vyBt71Wns3msLHn5qGnf+44XmaK41kfGPjqNzl85su912K6XPmzePr+2004rHXTbtwvx585q6eS1bA85Tl3Qr0AfYWFIlcBFwOTBG0mBgFnAMQERMlTQGmAYsA06LiKqvWaeQzaRpCzyQljo1SlCXNBQ4HhgNTEzJ3YFbJY2OiMtr2W8I2fQdKrr3oWLjHRqjeSXv3MEHsXz554y+/zkA5r77AdscfCEL3v+YXb7agzF/GMKuRw9f0Us//LSraNO6gpsuHUSfPbbl0Wdfa87mWyNZvHgx140cwYjrblw1s4ZvaHn52XtTacjnKyKOryVr/1q2Hw4MryF9EtBrTepurJ76YGCHiFhamCjpD8BUsk+sVRSefGi7y+nr5DjCCYftySH79eLgk65YkbZk6TIWvL8MgBdenc0ble/Sc4vOTJ42a8U2ny1Zxr3/nMJhfXZ0UM+pytmzeOutSo49qj8A8+bNZcDRR/G30bfTZdNNmTd37opt582dxyadO9dWlNWgLCc3yWisMfXPgc1qSO+a8qwGB3z9q5w9qC9Hn3Utiz/94vNw447tVrzhtuy2EVtvvgkzK9/lS21bs+nGHQAoLy+j3z7bM/1Nf+XOq57bbMv4J57hgUce5YFHHqVLl00ZfcedbLzJJvzXN7/Fg/ffx5IlS6isnM2sWW/Sa8evNXeTW5QmGFNvEo3VUz8LGCfpdb6Yf7k5sDVweiPV2aKMumwQ++7Wk403bMeMB3/JL0fczzknHkib1hXce032FFVNXfzGrlvz81MOZdny5SxfHpwxfDQLP/iEzp3ac8cfT6J1qwrKy8v453P/4ro7nmzmI7OGMvSnP2HScxNZtGghB3xrP0457QyO+s4xNW679dY9ObDfwRx5+CGUl5dz/gUXUl5eXuO2VrMSj9VFU2PNlpBUBvQmO1Eqsuk5zxWcAKjTujr8YnVb+NyVzd0EK0HrVaz9ac5thz5UdMyZ/uuDSvYjoNFmv0TE58CExirfzKwh5aWn7l+UmpmRnxOlDupmZjiom5nliodfzMxypNSnKhbLQd3MDAd1M7NcyUlMd1A3MwOfKDUzyxUPv5iZ5UhOYrqDupkZuKduZpYrOYnpDupmZuCeuplZrnj2i5lZjuSko+6gbmYGHn4xM8uVnMR0B3UzM3BP3cwsVxzUzcxyxLNfzMxyJCcddQd1MzPw8IuZWa7kJKY7qJuZAZTlJKo7qJuZ4ROlZma5kpOYTllzN8DMrBRIKnopoqwfS5oq6RVJt0paT1InSY9Iej3937Fg+2GSZkiaLumgtTmOWnvqkv4XiNryI+LMtanYzKyUNNSQuqRuwJnA9hGxWNIYYACwPTAuIi6XdB5wHjBU0vYpfwdgM+AfkraJiOX1qb+u4ZdJ9SnQzKwlEg06/lIBtJW0FFgfeBsYBvRJ+aOA8cBQoD8wOiI+A2ZKmgH0Bp6pb8U1iohRhY8lfSkiPq5PJWZmpW5NxtQlDQGGFCSNjIiRABHxlqTfAbOAxcDDEfGwpC4RMSdtM0dS57RvN2BCQVmVKa1+x1FE4/eWNA14NT3eSdLV9a3QzKwUlZWp6CUiRkbE7gXLyKpy0lh5f2ArsuGUL0n6Xh1V1/RxUuvQ92qPo4ht/ggcBLwHEBEvAfvVt0Izs1JUJhW9rEZfYGZEvBMRS4E7ga8D8yR1BUj/z0/bVwI9CvbvTjZcU7/jKGajiJhdLaleA/hmZqVKKn5ZjVnAXpLWVzZVZn+ykY6xwMC0zUDg7rQ+FhggqY2krYCewMT6Hkcx89RnS/o6EJJak53VfbW+FZqZlaKGuvZLRDwr6Q5gMrAMeAEYCbQDxkgaTBb4j0nbT00zZKal7U+r78wXKC6onwz8iWzg/i3gIeC0+lZoZlaKGvIqARFxEXBRteTPyHrtNW0/HBjeEHWvNqhHxLvACQ1RmZlZqSrPybVfipn98mVJ90h6R9J8SXdL+nJTNM7MrKk05C9Km1MxJ0pvAcYAXcmm59wO3NqYjTIza2plKn4pZcUEdUXEXyNiWVpuZi3mUJqZlaK89NTruvZLp7T6WLpOwWiyYH4ccF8TtM3MrMmUeKwuWl0nSp8nC+JVh3pSQV4Av2ysRpmZNbVS74EXq65rv2zVlA0xM2tO5aU+WF6kom6SIakX2WUj16tKi4i/NFajzMyaWj5CehFBXdJFZJeL3B64HzgYeBJwUDez3MjLPUqLmf1yNNmvoOZGxInATkCbRm2VmVkTa8BrvzSrYoZfFkfE55KWSepAdmUx//jIzHIl9ydKC0yStCFwHdmMmI9YiyuImZmVopzE9KKu/XJqWh0h6UGgQ0S83LjNMjNrWrmf/SJp17ryImJy4zTJzKzprQvDL7+vIy+AbzVwW1ayYOKVjVm8tVAde5/Z3E2wErR48hVrXUZRdwxqAer68dE3m7IhZmbNaV3oqZuZrTNyMqTuoG5mBuvAiVIzs3VJTmJ6UXc+kqTvSbowPd5cUu/Gb5qZWdPJyy9KiznhezWwN3B8evwhcFWjtcjMrBmUSUUvpayY4Zc9I2JXSS8ARMRCSa0buV1mZk0q91MaCyyVVE66hZ2kTYDPG7VVZmZNrMQ74EUrJqhfAdwFdJY0nOyqjRc0aqvMzJrYOjP7JSL+Jul5ssvvCjgiIl5t9JaZmTWhnMT0om6SsTnwCXBPYVpEzGrMhpmZNaVSPwFarGKGX+7jixtQrwdsBUwHdmjEdpmZNamcxPSihl92LHycrt54UqO1yMysGawzwy/VRcRkSXs0RmPMzJqLcnLr6WLG1H9S8LAM2BV4p9FaZGbWDCpyMlG9mMNoX7C0IRtj79+YjTIza2qSil6KKGtDSXdIek3Sq5L2ltRJ0iOSXk//dyzYfpikGZKmSzpobY6jzp56+tFRu4g4Z20qMTMrdQ08pv4n4MGIODr9An994HxgXERcLuk84DxgqKTtgQFkk082A/4haZuIWF6fimvtqUuqSIXWels7M7O8aKgLeknqAOwH3AAQEUsiYhHZCMeotNko4Ii03h8YHRGfRcRMYAZQ74sm1tVTn0gW0F+UNBa4Hfi4KjMi7qxvpWZmpWZN5qlLGgIMKUgaGREj0/qXyc47/lnSTsDzwI+ALhExByAi5kjqnLbvBkwoKKsypdVLMbNfOgHvkd2TtGq+egAO6maWG+VrcKI0BfCRtWRXkHWIz4iIZyX9iWyopTY1fZpE8a1ZtfLadE4zX17hi2C+1hWamZWisoab0lgJVEbEs+nxHWRBfZ6krqmX3hWYX7B9j4L9uwNv17fyuj6byoF2aWlfsF61mJnlRkONqUfEXGC2pG1T0v7ANGAsMDClDQTuTutjgQGS2kjaCuhJNvxdL3X11OdExCX1LdjMrCVp4NkvZwB/SzNf3gBOJOtEj5E0GJgFHAMQEVMljSEL/MuA0+o78wXqDur5+HmVmVkRGvKCXhHxIrB7DVn717L9cGB4Q9RdV1CvsXIzszzK/QW9ImJBUzbEzKw5rTM3yTAzWxfk5NIvDupmZkBR13RpCRzUzczIz8wQB3UzM9at29mZmeVePkK6g7qZGQBlnv1iZpYfnv1iZpYjnv1iZpYj+QjpDupmZoB76mZmuVLuoG5mlh/5COkO6mZmwDpwlUYzs3VJA97Orlk5qJuZ4Z66mVmuyD11M7P88OwXM7McyUlMd1A3MwMHdTOzXPGYuplZjuTkyrsO6mZm4DsfmZnlSl6GX/JyXfgW7aILhvHN/fbmO0d8e5W8UX++gZ17bcvChQsAWLp0KRecP5SjjzyMIw87mBuuu7apm2uNZMRF3+U//xjOpDHnrUi79Kz+vPj3nzHxtqHc9rvBbNCu7Ur79Ni0I+88+VvO+u9vrUh7aOQZvHTnz5hw67lMuPVcNunYrsmOoSUrU/FLKXNQLwGHH3EUV4+4fpX0uXPmMOGZp+nadbMVaY88/CBLlyzhjrvu4ZYxd3LH7bfx1luVTdlcayR/vedZ+p9+zUpp4yZMZ7djL6P3cb/m9VnvcM4PDlgp/zdnH8nDT01bpawTf/YX9jr+N+x1/G94Z+FHjdruvNAa/CtlDuolYLfd96DDBhuskv6731zGWT85Z6W5VpJYvHgxy5Yt47PPPqVVq1a0a+eeWB48NfnfLHj/k5XSxk14jeXLPwdg4pQ36dZ5wxV5h/XZkZlvvce0N+Y2ZTNzSyp+KWUO6iVq/GPj2KRzZ7bdbruV0vsecBBt27blgG9+g34HfJPvD/oBG2ywYfM00prU9/vvxUNPZ73y9ddrzdmD+jL82gdq3Pbai09gwq3nct4PD2rKJrZoWoOllDV5UJd0Yh15QyRNkjTphutHNmWzSsrixYu5fuQITj39R6vkvTLlZcrKy3j40Se4/8Fx/HXUjVTOnt0MrbSmdO7gA1m+bDmj758EwM9PPpj//dt4Pl68ZJVtT/zZX9jjuMvpO/hP7LPLV/juoXs0dXNbpHKp6KWUNcfsl18Af64pIyJGAiMBFi8lmrJRpaRy9izeequSY7/TH4D58+Zy/DFHcfPo23ng/nvZZ599adWqFZ022oidd96VqVOn0L1Hj2ZutTWWE77dm0P23YGDT75yRdoeO27JkX13ZviPDmeD9m35/PPg0yVLGXHbE7z9zvsAfPTJZ9z24CT26LUFt9z3XHM1v+Vo4FgtqRyYBLwVEd+W1Am4DdgSeBM4NiIWpm2HAYOB5cCZEfFQfettlKAu6eXasoAujVFnnvTcZlsee/yZFY8PPvBb3HLbHXTs2ImuXbsyceKzHHpYfz5dvJgpL7/ECf89sBlba43pgK9/lbMH9eXAH17B4k+XrkjvO/hPK9Z/dtLBfPzJZ4y47QnKy8vYsH1b3lv0MRUVZRyyby8efXZ6czS9xWmEE6A/Al4FOqTH5wHjIuJySeelx0MlbQ8MAHYANgP+IWmbiFhen0obq6feBTgIWFgtXcDTjVRni3XeOT9h0nMTWbRoIQfuvx+nnHoGR37nmBq3Pe74E7jwgmHZ9McIDj/iKLbZdrsat7WWZdSlA9l3t63ZeMN2zHjgEn454n7O+cEBtGlVwb3XnApkJ0vPvHRMrWW0aVXB2KtOpVVFGeVlZTz27HRuvMt/csVoyFEVSd2BQ4HhwE9Scn+gT1ofBYwHhqb00RHxGTBT0gygN/AM9aCIhh/lkHQD8OeIeLKGvFsi4rurK2NdHn6x2nXa88zmboKVoMWTr1jrkPzcG+8XHXN6f2XDk4AhBUkj0/AxAJLuAC4D2gM/TcMviyJiw4JtFkZER0lXAhMi4uaUfgPwQETcUZ/jaJSeekQMriNvtQHdzKzJrcHHQuH5v1WKkb4NzI+I5yX1qWfN9e7U+jIBZmY06LVf9gEOl3QIsB7QQdLNwDxJXSNijqSuwPy0fSVQONOhO/B2fSv3PHUzMxpunnpEDIuI7hGxJdkJ0Ecj4nvAWKBqVsNA4O60PhYYIKmNpK2AnsDE+h6He+pmZtAUvyq6HBgjaTAwCzgGICKmShoDTAOWAafVd+YLOKibmQGNc5XGiBhPNsuFiHgP2L+W7YaTzZRZaw7qZmaU/jVdiuWgbmaGg7qZWa6U+iV1i+WgbmaGe+pmZrmSk5juoG5mBuQmqjuom5nhMXUzs1wp9RtKF8tB3cwMPPxiZpYnHn4xM8sRT2k0M8uRnMR0B3UzMyA3Ud1B3cyMBr1JRrNyUDczIzcddQd1MzMgN1HdQd3MDE9pNDPLlZwMqTuom5mBg7qZWa54+MXMLEfcUzczy5GcxHQHdTMzcE/dzCxn8hHVHdTNzPBNMszMcsXDL2ZmOeIpjWZmeZKPmO6gbmYGuYnpDupmZpCfMfWy5m6AmVkpkFT0sppyekh6TNKrkqZK+lFK7yTpEUmvp/87FuwzTNIMSdMlHbQ2x+GgbmZGNvxS7LIay4CzI+KrwF7AaZK2B84DxkVET2BcekzKGwDsAPQDrpZUXt/jcFA3MyMbfil2qUtEzImIyWn9Q+BVoBvQHxiVNhsFHJHW+wOjI+KziJgJzAB61/c4HNTNzMimNBb9TxoiaVLBMqTGMqUtgV2AZ4EuETEHssAPdE6bdQNmF+xWmdLqxSdKzcxYsxOlETESGFl3eWoH/B04KyI+qGMsvqaMKL41K3NP3cyMhht+ycpSK7KA/reIuDMlz5PUNeV3Bean9EqgR8Hu3YG363scDupmZqzZ8Eud5WRd8huAVyPiDwVZY4GBaX0gcHdB+gBJbSRtBfQEJtb3ODz8YmZGg85T3wf4b2CKpBdT2vnA5cAYSYOBWcAxABExVdIYYBrZzJnTImJ5fSt3UDczo+F+URoRT9ZR3P617DMcGN4Q9Tuom5lBbq4T4KBuZoav0mhmliu+SYaZWZ44qJuZ5YeHX8zMciQvl95VRL1/jWpNRNKQ9LNksxX8vrCa+BelLUONFwuydZ7fF7YKB3UzsxxxUDczyxEH9ZbB46ZWE78vbBU+UWpmliPuqZuZ5YiDuplZjjiolzhJ/SRNlzRD0nnN3R5rfpJulDRf0ivN3RYrPQ7qJUxSOXAVcDCwPXC8pO2bt1VWAm4C+jV3I6w0OaiXtt7AjIh4IyKWAKOB/s3cJmtmEfE4sKC522GlyUG9tHUDZhc8rkxpZmY1clAvbTVdYshzUM2sVg7qpa0S6FHwuDvwdjO1xcxaAAf10vYc0FPSVpJaAwOAsc3cJjMrYQ7qJSwilgGnAw8BrwJjImJq87bKmpukW4FngG0lVUoa3NxtstLhywSYmeWIe+pmZjnioG5mliMO6mZmOeKgbmaWIw7qZmY54qBudZK0XNKLkl6RdLuk9deirJskHZ3Wr6/r4mSS+kj6ej3qeFPSxsWmV9vmozWs62JJP13TNpo1Jgd1W53FEbFzRPQClgAnF2amK0musYj4YURMq2OTPsAaB3WzdZ2Duq2JJ4CtUy/6MUm3AFMklUv6raTnJL0s6SQAZa6UNE3SfUDnqoIkjZe0e1rvJ2mypJckjZO0JdmHx4/Tt4R9JW0i6e+pjuck7ZP23UjSw5JekHQtNV8vZyWS/k/S85KmShpSLe/3qS3jJG2S0r4i6cG0zxOStmuQZ9OsEVQ0dwOsZZBUQXZd9wdTUm+gV0TMTIHx/YjYQ1Ib4ClJDwO7ANsCOwJdgGnAjdXK3QS4DtgvldUpIhZIGgF8FBG/S9vdAvxPRDwpaXOyX9l+FbgIeDIiLpF0KLBSkK7FD1IdbYHnJP09It4DvgRMjoizJV2Yyj6d7AbPJ0fE65L2BK4GvlWPp9Gs0Tmo2+q0lfRiWn8CuIFsWGRiRMxM6QcCX6saLwc2AHoC+wG3RsRy4G1Jj9ZQ/l7A41VlRURt1wnvC2wvreiId5DUPtVxVNr3PkkLizimMyUdmdZ7pLa+B3wO3JbSbwbulNQuHe/tBXW3KaIOs2bhoG6rszgidi5MSMHt48Ik4IyIeKjadoew+ksFq4htIBsq3DsiFtfQlqKvdSGpD9kHxN4R8Ymk8cB6tWweqd5F1Z8Ds1LlMXVrCA8Bp0hqBSBpG0lfAh4HBqQx967AN2vY9xngvyRtlfbtlNI/BNoXbPcw2VAIabud0+rjwAkp7WCg42raugGwMAX07ci+KVQpA6q+bXyXbFjnA2CmpGNSHZK002rqMGs2DurWEK4nGy+fnG6GfC3Zt8C7gNeBKcA1wD+r7xgR75CNg98p6SW+GP64Bziy6kQpcCawezoRO40vZuH8AthP0mSyYaBZq2nrg0CFpJeBXwITCvI+BnaQ9DzZmPklKf0EYHBq31R8S0ErYb5Ko5lZjrinbmaWIw7qZmY54qBuZpYjDupmZjnioG5mliMO6mZmOeKgbmaWI/8fJluCwFZdos4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_svm = svm.SVC(random_state=42)\n",
    "model_svm.fit(x_train_twitter, y_train)\n",
    "predictions = model_svm.predict(x_test_twitter)\n",
    "evaluate(y_test, predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "bc3cd0d1-64a3-4769-86cc-c4b1a55341d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.8944    0.8995    0.8969      1393\n",
      "         1.0     0.8989    0.8938    0.8963      1393\n",
      "\n",
      "    accuracy                         0.8966      2786\n",
      "   macro avg     0.8966    0.8966    0.8966      2786\n",
      "weighted avg     0.8966    0.8966    0.8966      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9181f0a1-050d-4811-a09d-e69c0a7f5c40",
   "metadata": {},
   "source": [
    "#### Using **Logistic Regression** model with **hyperparameter tuning** (1,000 iterations to achieve convergence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5da00949-d538-4b6d-b805-2318d87e014c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best l1_ratio: 0.1\n",
      "Best C: 3\n",
      "Best penalty: elasticnet\n",
      "Best solver: saga\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_lr = {\"C\":[1, 3, 5, 7], 'l1_ratio':['none', 0, 0.1, 0.3, 0.5], 'penalty':['l2', 'elasticnet'],\n",
    "       'solver':['lbfgs', 'saga']}\n",
    "\n",
    "model_lr_1000 = LogisticRegression(max_iter=1000, random_state=42)\n",
    "\n",
    "logreg_cv_twitter=GridSearchCV(model_lr_1000, grid_lr, scoring='f1', cv=2)\n",
    "logreg_cv_twitter.fit(x_train_twitter, y_train)\n",
    "\n",
    "print('Best l1_ratio:', logreg_cv_twitter.best_estimator_.get_params()['l1_ratio'])\n",
    "print('Best C:', logreg_cv_twitter.best_estimator_.get_params()['C'])\n",
    "print('Best penalty:', logreg_cv_twitter.best_estimator_.get_params()['penalty'])\n",
    "print('Best solver:', logreg_cv_twitter.best_estimator_.get_params()['solver'])\n",
    "\n",
    "predictions = logreg_cv_twitter.predict(x_test_twitter)\n",
    "evaluate(y_test, predictions )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c5b41c3c-382c-4b57-9189-67997249a08d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.9024    0.8959    0.8991      1393\n",
      "         1.0     0.8967    0.9031    0.8999      1393\n",
      "\n",
      "    accuracy                         0.8995      2786\n",
      "   macro avg     0.8995    0.8995    0.8995      2786\n",
      "weighted avg     0.8995    0.8995    0.8995      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96b30608-f509-4c9e-814a-cfd2f8f5aa55",
   "metadata": {},
   "source": [
    "#### Using **SVM** model with **hyperparameter tuning**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8cb1513a-5be0-42e0-88cc-de5469904c91",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best KERNEL: rbf\n",
      "Best C: 140\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid_svm = [\n",
    "    {'kernel':('linear', 'rbf')},\n",
    "    {'C':[1, 40, 80, 120, 140, 160]},\n",
    "]    \n",
    "\n",
    "model_svm = svm.SVC(random_state=42)\n",
    "\n",
    "\n",
    "svm_model_cv_twitter = GridSearchCV(model_svm, grid_svm, n_jobs=-1)\n",
    "\n",
    "svm_model_cv_twitter.fit(x_train_twitter, y_train)\n",
    "\n",
    "print('Best KERNEL:', svm_model_cv_twitter.best_estimator_.get_params()['kernel'])\n",
    "print('Best C:', svm_model_cv_twitter.best_estimator_.get_params()['C'])\n",
    "\n",
    "predictions = svm_model_cv_twitter.predict(x_test_twitter)\n",
    "evaluate(y_test, predictions )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "30e92888-21b8-4255-99e9-9b75006418c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.9341    0.9763    0.9547      1393\n",
      "         1.0     0.9752    0.9311    0.9526      1393\n",
      "\n",
      "    accuracy                         0.9537      2786\n",
      "   macro avg     0.9546    0.9537    0.9537      2786\n",
      "weighted avg     0.9546    0.9537    0.9537      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de288994-9034-47c7-b829-053fc3fe24b3",
   "metadata": {},
   "source": [
    "### **glove-wiki-gigaword-300**: pre-trained on Wikipedia and Gigaword 5 Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "bd839822-a5e2-4084-9624-b5e62a0b21da",
   "metadata": {},
   "outputs": [],
   "source": [
    "glove_wiki = EmbeddingTransformer('glove-wiki-gigaword-300')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6a66a8ce-82e6-400b-b6c1-0d668c0946b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train_wiki = glove_wiki.transform(corpus_train) \n",
    "x_test_wiki = glove_wiki.transform(corpus_test) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdfc5007-40ff-40f5-8074-acc13d1a5f62",
   "metadata": {},
   "source": [
    "#### Using **Logistic Regression** model with the **default hyperparameters**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c3db7c33-0b07-4b67-88b6-00ccca822a06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAEWCAYAAACZnQc8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAlsklEQVR4nO3de7wVVd3H8c/3gOIFVFAhBLykaKmVpeKlJJ60vCZWXjAtMHtIUynTRH1MTUPpquU1UpMyRTQvmDcMJbJUbl6RSBIVggQEFJVUDr/nj1kHh+M5h83mnLP3Gb5vXvNiZs3MWmv27PPba69Ze0YRgZmZFUNNpStgZmbNx0HdzKxAHNTNzArEQd3MrEAc1M3MCsRB3cysQBzU2zhJJ0t6VdKbkjZvpTK7SZogaamkn69FPudKuq4561Ypko6TNLbMfT8t6YV0Do9o5qq1Okk3SvpRpeuxrmqTQV3SS5IOqJc2SNKjzZR/SNqhOfJqSZLWA34BfCEiOkbEa61U9GBgIbBJRJxRbiYRcUlEfLP5qtX8JG2b3g/tm9ouIv4QEV8os5iLgCvTObyrzDwAkHR/+nB4U9J7kt7NLV9bRn4XSrppbeq0mvzHS6rq90Bb0+Qb1apeN2ADYForl7sN8Hz4l2sASGofEcvXIottKPMc1i87Ig7OrbsRmBMR561F3ayNaZMt9VJI2krSHyUtkDRL0pDcuj6SHpO0RNI8SVdKWj+tm5A2ezq1bo6R1E/SHElnSZqf9jlC0iGS/ilpkaRzS8k/rQ9JQyS9KGmhpJ9KavBcSOog6XJJc9N0eUrbEZiRNlsi6eFG9v+MpL+nusyWNCilbyrpd+n1eVnSeXV1qPvWI+lnkhan1+/gtO5GYCBwVnp9Dqj/dbvu9cotD5X079RdM0PS/il9lVagpMMlTUt1HS/po7l1L0k6U9Izkl6XdKukDRo55kGS/ibpspTXi5L2Temz0zkcmNv+UElPSnojrb8wl13d+2FJOt596uW/CLhQuW+KqayFknql5U+kenykgbr+C/gwcE/Kv0N6745J76uZkv43t/2Fkm6XdJOkN4BBDb0Gjbwuh0l6KtXl75I+3tQ5knQQcC5wTKrb043k+0lJU9O+t5I1NOrWdZb0p/Q+W5zme6Z1w4D9gCtT/lem9F+m8/CGpCmS9iv1GA2IiDY3AS8BB9RLGwQ8muZrgCnA+cD6ZH80LwIHpvW7A3uTfVPZFpgOfDeXVwA75Jb7ActTfusB/wssAG4GOgG7AP8FPrwG+T8CdAG2Bv4JfLORY70IeBzoCmwJ/B24OK3bNuXVvpF9twaWAsemem8O7JbW/Q64O9V/21SHE3Ov5XvpONsBJwNzAaX1NwI/ypVTf7kfWQsRYCdgNrBVrs7bp/kLgZvS/I7AW8DnU13PAmYC6+fO+URgq/S6TQdOauS4B6XzdUKq/4+AV4CrgA7AF9Lr0jFX34+RvW8+DrwKHNHYa5zL/7R0jjck9/5L2wwDHk7rngFOLfX9DPwFuJosOO5G9l7bP/eavQcckeq7YRP5rjwvwKeA+cBe6TUZmMrtUOo5aqSM9YGXgdPTeTsy1a+u3M2BrwAbkb3XbgPuyu0/nnrvfeD4tF974AzgP8AGlY47bWWqeAXKqnT2ZnwTWJKb3ub9oL4X8Eq9fc4BfttIft8F7swtNxTUlwHt0nKntM1euW2m1AWCEvM/KLf8bWBcI/v+Czgkt3wg8FKa35amg/o5+XJz6e2Ad4Cdc2nfAsan+UHAzNy6jVI5H0rLK4NFI8v9eD+o70AWTA4A1qtXj5UBA/gBMDq3rgb4N9Avd86Pz63/CXBtI8c9CHght/yxVP9uubTXSB9wDex/OXBZY69xyr/++2sQqwb19dJ74lngAdIHYhPv5wPSfC+gFuiUW38pcGPuNZtQ4t/JyvMCXENqDOTWzwA+W+o5aqSMvuQ+8FPa3/Pvh3rb7wYszi2Pp5EGTW6bxcAnSjlmT9Gmu1+OiIjN6iaywFhnG2Cr9DVziaQlZF8juwFI2jF9DfxP+gp7CbDFasp7LSJq0/yy9P+rufXLgI5rkP/s3PzLZC3QhmyV1peybX29yD4U6tuC91tY+Xx75Jb/UzcTEW+n2Y4llrtSRMwk+1C7EJgvaZSkhuq/ynFGxAqy16jBOpF9iDdVn/rnhoho7HztJemR1EXwOnASq38/zG5qZUS8RxZUdwV+Hik6lWArYFFELM2l1T83TZbdiG2AM+r9TfQia52Xeo4aq++/6x3fyvMoaSNJv1bWxfcGWXfWZpLaNZahpDMkTU/dbEuATVn9+bCkLQf1pswGZuWDfkR0iohD0vprgH8AvSNiE7KAr2Ysv5T8e+XmtyZr7TRkLtkfZCnb1jcb2L6B9IVkX5Hr5/vvEvOt7y2y1nydD+VXRsTNEfGZVF4AP24gj1WOU5LIXqNy67QmbgbGAL0iYlPgWt4/X40F4yaDtKQewAXAb4GfS+pQYl3mAl0kdcql1T835Vygng0Mq/c3sVFE3AJNnqPVlTUP6JHOV76+dc4g697ZK/0t9E3pDb6+qf98KHA00Dk12F6nef8+C62oQX0i8Ea6+LOhpHaSdpW0Z1rfCXgDeDNdvDq53v6vkvXDl2t1+QN8P11E6gV8B7i1kbxuAc6TtKWkLcj69UsdYvYH4ABJR0tqL2lzSbulbxyjgWGSOknaBvjeGuRb31PAIZK6SPoQWasPAEk7SfpcCmr/JWsh1zaQx2jg0HSBbj2yYPAO2Vf5ltaJrHX8X0l9gK/m1i0AVrAG74cU4G4ErgdOJAt8F5eyb0TMJjvmSyVtkC5mnkh2LtfGb4CT0rcSSdpY2QXiTqs5R68C26qRC/nAY2TXF4ak99iXgT659Z1SfkskdSH7oMur/7fWKeW3AGgv6Xxgk7KPeh1UyKCegtYXyfrvZpG1TK8j+xoHcCbZH+5Ssjd7/YB6ITAyfU09uowqrC5/yC5STiELiPeSBYCG/AiYTHax7VlgakpbrYh4BTiELEAuSmV9Iq0+jayF/SLwKFlr9YZS8m3A74GnyfqGx7Lq8XYAhpOdg/+QXfA9t97+RMQMsgtkV6Rtvwh8MSLeLbNOa+LbwEWSlpJ9aI7O1ettsouef0vvh71LyG8IWVffD1K3xAnACWswiuNYsr78ucCdwAUR8VCpB9OQiJhMduH7SrI+6pm8P3KmqXN0W/r/NUlTG8j3XeDLKa/FwDHAHblNLie7WLyQ7IL/A/Wy+CVwZBoZ8yvgQeB+sgv3L5N9yJTT3bTOqhvNYK1IUpB1zcysdF3MrFgK2VI3M1tXOaibmRWIu1/MzArELXUzswKp2ht6bfjJU/0Vwj5g8aQrK10Fq0IbtF/7cexrEnOWPXll1Y6br9qgbmbWqhodit+2OKibmQGoahvfa8RB3cwM3FI3MysUt9TNzAqkptEbR7YpDupmZuDuFzOzQilI90sxPprMzNaWakqfVpeVdIOyZ+E+l0v7qaR/KHvO7p2SNsutO0fZ82hnSDowl767pGfTul/Vu299gxzUzcwga6mXOq3ejcBB9dIeAnaNiI+T3Vr4nKxY7QwMIHvW8UHA1bknQ10DDAZ6p6l+nh/goG5mBs3aUo+ICWTPMMinjY2I5WnxcaBnmu8PjIqIdyJiFtm97vtI6g5sEhGPpfvy/47sgeNNclA3M4Ns9EuJk6TBkibnpsFrWNo3yB4GAtnzZ/MPApmT0nqk+frpTfKFUjMzWKPRLxExAhhRVjHS/5E9sq/uEYUN9edEE+lNclA3MwOoafnRL5IGAocB+8f79z2fw6oPou9J9ijDObzfRZNPb5K7X8zMoFn71BvMXjoIGAocnp59W2cMMEBSB0nbkV0QnRgR84ClkvZOo16+TvZs4ya5pW5mBs06Tl3SLUA/YAtJc4ALyEa7dAAeSiMTH4+IkyJimqTRwPNk3TKnRERtyupkspE0G5L1wd/Pajiom5lBs94mICKObSD5+ia2HwYMayB9MrDrmpTtoG5mBr5NgJlZoRTkNgEO6mZm4Ja6mVmhuKVuZlYgbqmbmRWIH5JhZlYgbqmbmRWI+9TNzArELXUzswJxS93MrEDcUjczKw7VOKibmRVGCc90bhMc1M3MoOHnDLVBDupmZrilbmZWKA7qZmYFUuMLpWZmBVKMhrqDupkZuPvFzKxQHNTNzArEQd3MrEAc1M3MCkQ1DupmZoXhlrqZWYE4qJuZFUkxYrqDupkZuKVuZlYoDupmZgXie7+YmRVJMRrqFOOjycxsLUkqeSohrxskzZf0XC6ti6SHJL2Q/u+cW3eOpJmSZkg6MJe+u6Rn07pfqYTCHdTNzGjeoA7cCBxUL+1sYFxE9AbGpWUk7QwMAHZJ+1wtqV3a5xpgMNA7TfXz/AAHdTMzmjeoR8QEYFG95P7AyDQ/Ejgilz4qIt6JiFnATKCPpO7AJhHxWEQE8LvcPo1yUDczI7tNQMmTNFjS5Nw0uIQiukXEPID0f9eU3gOYndtuTkrrkebrpzfJF0or5NoLjuPgvruyYNFS9jjqEgAu+e4RHNJ3V959r5ZZcxYy+IKbeP3NZWzdvQtP3XEe/3x5PgATn32JIcNGAbBe+3ZcdvbR9N2jNytWrODCq/7EXeOeqtRhWTM6/7xzmPCX8XTpsjl33P0nAK781eWMf2QcNaqh8+abc/GwS+natRv3/mkMI2+4fuW+//znDEbddicf+ehHK1X9NmdNhjRGxAhgRHMV3VARTaQ3yS31Cvn9PY/T/5SrVkkb9/g/2P2oS+hzzKW88PJ8vv+NL6xc9+Kchew9YDh7Dxi+MqADDP3mgSxYtJSPH3ERn/zKMP465YVWOwZrWf2P+DLX/Pq6VdIGfeOb3H7nPYy+4276frYfv74mew8detjhjL7jbkbfcTfDhv+ErXr0cEBfQ83cp96QV1OXCun/+Sl9DtArt11PYG5K79lAepMc1Cvkb1P/xaLX314lbdzj/6C2dgUAE5+dRY9um602n4H99+GnN4wFICJ4bclbzV5Xq4zd99iTTTbddJW0jh07rpz/77JlDQaY+++7l4MPOazF61c0rRDUxwAD0/xA4O5c+gBJHSRtR3ZBdGLqolkqae806uXruX0a1WLdL5I+QnYBoAfZV4a5wJiImN5SZRbJ1/vvw+1jp65c3rbH5jx2y1CWvvVffnjVn/jbk/9i044bAnDBKYex3+69mTVnAacPv435i5ZWqtrWCq745WXcM+YuOnbsxHW//d0H1j/4wH1cfsXVFahZG9eM49Ql3QL0A7aQNAe4ABgOjJZ0IvAKcBRAREyTNBp4HlgOnBIRtSmrk8lG0mwI3J+mJrVIS13SUGAU2cs0EZiU5m+RdHYT+628+LB84bSWqFqbcNaJB1Jbu4JR900C4D8L32DHg89nn2N/zNCf38GNlwyi08Yb0L59DT0/1JnHnnqRfb/6Y5545iUuPf1LFa69tbTTvnM6Y8f9hUMP+yKjbr5plXXPPPM0G2ywIb1771ih2rVdzTz65diI6B4R60VEz4i4PiJei4j9I6J3+n9RbvthEbF9ROwUEffn0idHxK5p3alpFEyTWqr75URgz4gYHhE3pWk40Ceta1BEjIiIPSJij/Zb7NJCVatux31xLw7puyuD/u/GlWnvvrecRa9n3SpPTp/Ni3MW0nubrry25C3eWvYOdz/8NAB3PDSV3T7aq6FsrYAOPvQw/vzQ2FXSHrzvXg4+5NAK1ahtq6lRyVM1a6mgvgLYqoH07mmdNeDz+36UMwYdwJHf/TXL/vveyvQtOndc+Ubatsfm7LD1lsyasxCA+yY8R989egPQr89O/OPFea1fcWs1L7/80sr58Y88zHbbfXjl8ooVKxg79gEOOthBvRyt0KfeKlqqT/27wDhJL/D++MutgR2AU1uozDZl5KWD2G/33myxWUdmPnAxF197H98/4Qt0WL89f7ome4nqhi5+5lM78IOTD2V5bS21tcFpw0ax+I3sIut5v7yL6380kJ+e+RUWLn6Tb114U1PFWhsy9MzvMXnSRJYsWcznP9eXk085jUcnTOCll2ZRUyO6d+/BeRf8cOX2UyZPolu3D9Gzl7+tlaPKY3XJVEIXTXkZSzVk3S09yPrT5wCTchcAmrThJ09tmYpZm7Z40pWVroJVoQ3ar/1lzp2GPlhyzJnx4wOr9iOgxUa/RMQK4PGWyt/MrDkVpaXuX5SamUHVXwAtlYO6mRkO6mZmheLuFzOzAqn2oYqlclA3M8NB3cysUAoS0x3UzczAF0rNzArF3S9mZgVSkJjuoG5mBm6pm5kVSkFiuoO6mRm4pW5mVige/WJmViAFaag7qJuZgbtfzMwKpSAx3UHdzAzcUjczKxQHdTOzAvHoFzOzAilIQ91B3cwM3P1iZlYoBYnpDupmZgA1BYnqDupmZvhCqZlZoRQkplNT6QqYmVUDSSVPJeR1uqRpkp6TdIukDSR1kfSQpBfS/51z258jaaakGZIOXJvjaLSlLukKIBpbHxFD1qZgM7Nq0lxd6pJ6AEOAnSNimaTRwABgZ2BcRAyXdDZwNjBU0s5p/S7AVsCfJe0YEbXllN9U98vkcjI0M2uLRLP2v7QHNpT0HrARMBc4B+iX1o8ExgNDgf7AqIh4B5glaSbQB3is3IIbFBEj88uSNo6It8opxMys2q1Jn7qkwcDgXNKIiBgBEBH/lvQz4BVgGTA2IsZK6hYR89I28yR1Tfv2AB7P5TUnpZV3HCVUfh9JzwPT0/InJF1dboFmZtWopkYlTxExIiL2yE0j6vJJfeX9ge3IulM2lnR8E0U39HHSaNf3ao+jhG0uBw4EXgOIiKeBvuUWaGZWjWqkkqfVOACYFRELIuI94A5gX+BVSd0B0v/z0/ZzgF65/XuSddeUdxylbBQRs+slldWBb2ZWraTSp9V4Bdhb0kbKhsrsT9bTMQYYmLYZCNyd5scAAyR1kLQd0BuYWO5xlDJOfbakfYGQtD7ZVd3p5RZoZlaNmuveLxHxhKTbganAcuBJYATQERgt6USywH9U2n5aGiHzfNr+lHJHvkBpQf0k4JdkHff/Bh4ETim3QDOzatScdwmIiAuAC+olv0PWam9o+2HAsOYoe7VBPSIWAsc1R2FmZtWqXUHu/VLK6JcPS7pH0gJJ8yXdLenDrVE5M7PW0py/KK2kUi6U3gyMBrqTDc+5DbilJStlZtbaalT6VM1KCeqKiN9HxPI03cRajKE0M6tGRWmpN3Xvly5p9pF0n4JRZMH8GODeVqibmVmrqfJYXbKmLpROIQvidYf6rdy6AC5uqUqZmbW2am+Bl6qpe79s15oVMTOrpHbV3lleopIekiFpV7LbRm5QlxYRv2upSpmZtbZihPQSgrqkC8huF7kzcB9wMPAo4KBuZoVRlGeUljL65UiyX0H9JyJOAD4BdGjRWpmZtbJmvPdLRZXS/bIsIlZIWi5pE7I7i/nHR2ZWKIW/UJozWdJmwG/IRsS8yVrcQczMrBoVJKaXdO+Xb6fZayU9AGwSEc+0bLXMzFpX4Ue/SPpUU+siYmrLVMnMrPWtC90vP29iXQCfa+a6rOK1J65oyeytjeq89+mVroJVoWWTL1vrPEp6YlAb0NSPj/6nNStiZlZJ60JL3cxsnVGQLnUHdTMzWAculJqZrUsKEtNLevKRJB0v6fy0vLWkPi1fNTOz1lOUX5SWcsH3amAf4Ni0vBS4qsVqZGZWATVSyVM1K6X7Za+I+JSkJwEiYrGk9Vu4XmZmrarwQxpz3pPUjvQIO0lbAitatFZmZq2syhvgJSslqP8KuBPoKmkY2V0bz2vRWpmZtbJ1ZvRLRPxB0hSy2+8KOCIiprd4zczMWlFBYnpJD8nYGngbuCefFhGvtGTFzMxaU7VfAC1VKd0v9/L+A6g3ALYDZgC7tGC9zMxaVUFiekndLx/LL6e7N36rxWpkZlYB60z3S30RMVXSni1RGTOzSlFBHj1dSp/693KLNcCngAUtViMzswpoX5CB6qUcRqfc1IGsj71/S1bKzKy1SSp5KiGvzSTdLukfkqZL2kdSF0kPSXoh/d85t/05kmZKmiHpwLU5jiZb6ulHRx0j4vtrU4iZWbVr5j71XwIPRMSR6Rf4GwHnAuMiYriks4GzgaGSdgYGkA0+2Qr4s6QdI6K2nIIbbalLap8ybfSxdmZmRdFcN/SStAnQF7geICLejYglZD0cI9NmI4Ej0nx/YFREvBMRs4CZQNk3TWyqpT6RLKA/JWkMcBvwVt3KiLij3ELNzKrNmoxTlzQYGJxLGhERI9L8h8muO/5W0ieAKcB3gG4RMQ8gIuZJ6pq27wE8nstrTkorSymjX7oAr5E9k7RuvHoADupmVhjt1uBCaQrgIxpZ3Z6sQXxaRDwh6ZdkXS2NaejTJEqvzQcLb0zXNPLlOd4P5mtdoJlZNappviGNc4A5EfFEWr6dLKi/Kql7aqV3B+bntu+V278nMLfcwpv6bGoHdExTp9x83WRmVhjN1aceEf8BZkvaKSXtDzwPjAEGprSBwN1pfgwwQFIHSdsBvcm6v8vSVEt9XkRcVG7GZmZtSTOPfjkN+EMa+fIicAJZI3q0pBOBV4CjACJimqTRZIF/OXBKuSNfoOmgXoyfV5mZlaA5b+gVEU8BezSwav9Gth8GDGuOspsK6g0WbmZWRIW/oVdELGrNipiZVdI685AMM7N1QUFu/eKgbmYGlHRPl7bAQd3MjOKMDHFQNzNj3XqcnZlZ4RUjpDuom5kBUOPRL2ZmxeHRL2ZmBeLRL2ZmBVKMkO6gbmYGuKVuZlYo7RzUzcyKoxgh3UHdzAxYB+7SaGa2LmnGx9lVlIO6mRluqZuZFYrcUjczKw6PfjEzK5CCxHQHdTMzcFA3MysU96mbmRVIQe6866BuZgZ+8pGZWaG4+8WazYU/OJcJE8bTpcvm3H7nPQBce/UV3PHH2+jcuQsApw45nf36fnblPvPmzeUr/Q/jpG+fwtcHnViRelvzuvb8ARz8mZ1ZsPhN9jjmJwBcMuSLHNJ3F959r5ZZcxYy+Ie38Pqb/2WPXbbmynOPBrILfMNGPMiY8c8CcOTnd+Osb3yedjU1PPC35/m/X91TsWNqS4rS/VKUh320aV/s/yWuuuY3H0g//msDufX2u7j19rtWCegAP/vJpXz6M/u1VhWtFfz+non0P23EKmnjnvgnux/zE/oc+1NeeGUB3z/hAACmzZzHp7/+C/Y+7mf0P20EV5x7FO3a1dBl04245DuHc8jJV7P7MT+ma5dO9NuzdyUOp83RGvyrZg7qVWD3PfZk0003LXn7R8b9mZ49e7H9Dju0YK2stf3tyRdZ9MZbq6SNe2IGtbUrAJj47Mv06LoZAMveeW9leocO7YnItt+ux+a88PICFi7J8nl44j854nMfb50DaOOk0qdq5qBexUbd8geO/vLhXPiDc3nj9dcBWPb22/z2ht/wrZNPqXDtrLV9/fC9ePDv01cu77nL1ky5dSiTR53FkEtvo7Z2Bf+avZCdtu3K1t07065dDYf325We3TpXsNZth9ZgqmatHtQlndDEusGSJkuafMN1IxrbbJ1w1NHHcs99DzHq9rvYYsst+cXPfgzANVdfwfFfG8RGG21c4RpaazrrGwdQW1vLqPunrEybNO0Vdj/mx3zm67/g+yfsT4f127Nk6TKGDL+dmy4dyLjfnMbLcxdRW1tbwZq3He2kkqdqVokLpT8EftvQiogYAYwAePvdui+U66bNt9hi5fyXv3IUQ049GYDnnn2GPz/0IJdf9lOWLl1KjWpYf/0ODPjq8ZWqqrWw4w7dk0M+swsHn3x1g+tnvDSft5a9yy7bd2fq9Nnc99dp3PfXaQB840v7ULtinf5TKl0zx2pJ7YDJwL8j4jBJXYBbgW2Bl4CjI2Jx2vYc4ESgFhgSEQ+WW26LBHVJzzS2CujWEmUWzYIF89lyy64APDzuz2y/Q3ax64aRf1i5zbVXX8FGG23kgF5gn9/nI5wx8HN8YfCVLHvnvZXp22zVhTmvLqG2dgVbf6gzO27TlZfnLgJgy84dWbD4TTbrtCGDj/w0x58zslLVb1Na4ALod4DpwCZp+WxgXEQMl3R2Wh4qaWdgALALsBXwZ0k7RkRZX7FaqqXeDTgQWFwvXcDfW6jMNuvss77HlEmTWLJkMQfu/1lOOuU0pkyayIx/TEcS3Xv04Lzzf1jpaloLGznsa+y3+w5ssdnGzLz3Ai4e8QDfH7Q/HdZrz5+uyr6pTXzuZYZcehv77vZhzhy4P+8tr2VFBN8ZfjuvvZ5dHP3ZmV/iY723AuDS68Yy85UFFTumtqQ5e1Uk9QQOBYYB30vJ/YF+aX4kMB4YmtJHRcQ7wCxJM4E+wGNllR0t0Msh6XrgtxHxaAPrbo6Ir64uj3W9+8Uatvm+31v9RrbOWTb5srUOyZNefL3kmNNn+82+BQzOJY1I3ccASLoduBToBJyZul+WRMRmuW0WR0RnSVcCj0fETSn9euD+iLi9nONokZZ6RDT6a5hSArqZWatbg4+F/PW/D2QjHQbMj4gpkvqVWXLZjVr/otTMjGa998ungcMlHQJsAGwi6SbgVUndI2KepO7A/LT9HKBXbv+ewNxyC/c4dTMzmm+cekScExE9I2JbsgugD0fE8cAYYGDabCBwd5ofAwyQ1EHSdkBvYGK5x+GWupkZtMavioYDoyWdCLwCHAUQEdMkjQaeB5YDp5Q78gUc1M3MgJa5S2NEjCcb5UJEvAbs38h2w8hGyqw1B3UzM6r/ni6lclA3M8NB3cysUKr9lrqlclA3M8MtdTOzQilITHdQNzMDChPVHdTNzHCfuplZoRTlwdMO6mZm4O4XM7MicfeLmVmBeEijmVmBFCSmO6ibmQGFieoO6mZmNOtDMirKQd3MjMI01B3UzcyAwkR1B3UzMzyk0cysUArSpe6gbmYGDupmZoXi7hczswJxS93MrEAKEtMd1M3MwC11M7OCKUZUd1A3M8MPyTAzKxR3v5iZFYiHNJqZFUkxYrqDupkZFCamU1PpCpiZVQOp9KnpfNRL0iOSpkuaJuk7Kb2LpIckvZD+75zb5xxJMyXNkHTg2hyHg7qZGSCp5Gk1lgNnRMRHgb2BUyTtDJwNjIuI3sC4tExaNwDYBTgIuFpSu3KPw0HdzIys+6XUqSkRMS8ipqb5pcB0oAfQHxiZNhsJHJHm+wOjIuKdiJgFzAT6lHscDupmZjRf98uqeWpb4JPAE0C3iJgHWeAHuqbNegCzc7vNSWllcVA3MyMb0ljyP2mwpMm5afAH8pM6An8EvhsRbzRZ9AdFucfh0S9mZqxZCzwiRgAjGs9L65EF9D9ExB0p+VVJ3SNinqTuwPyUPgfoldu9JzB3Daq+CrfUzcxo1tEvAq4HpkfEL3KrxgAD0/xA4O5c+gBJHSRtB/QGJpZ7HG6pm5nRrL8o/TTwNeBZSU+ltHOB4cBoSScCrwBHAUTENEmjgefJRs6cEhG15RbuoG5mRvPd+yUiHqXxQTL7N7LPMGBYc5TvoG5mRnF+UeqgbmYGhYnqDupmZvgujWZmheKHZJiZFYmDuplZcbj7xcysQIryODtFlH2LAWslkgannyWbreT3hTXEtwloGz5wsyAz/L6wBjiom5kViIO6mVmBOKi3De43tYb4fWEf4AulZmYF4pa6mVmBOKibmRWIg3qVk3SQpBmSZko6u9L1scqTdIOk+ZKeq3RdrPo4qFcxSe2Aq4CDgZ2BYyXtXNlaWRW4ETio0pWw6uSgXt36ADMj4sWIeBcYBfSvcJ2swiJiArCo0vWw6uSgXt16ALNzy3NSmplZgxzUq1tDtxjyGFQza5SDenWbA/TKLfcE5laoLmbWBjioV7dJQG9J20laHxgAjKlwncysijmoV7GIWA6cCjwITAdGR8S0ytbKKk3SLcBjwE6S5kg6sdJ1surh2wSYmRWIW+pmZgXioG5mViAO6mZmBeKgbmZWIA7qZmYF4qBuTZJUK+kpSc9Juk3SRmuR142Sjkzz1zV1czJJ/STtW0YZL0naotT0etu8uYZlXSjpzDWto1lLclC31VkWEbtFxK7Au8BJ+ZXpTpJrLCK+GRHPN7FJP2CNg7rZus5B3dbEX4EdUiv6EUk3A89Kaifpp5ImSXpG0rcAlLlS0vOS7gW61mUkabykPdL8QZKmSnpa0jhJ25J9eJyeviXsJ2lLSX9MZUyS9Om07+aSxkp6UtKvafh+OauQdJekKZKmSRpcb93PU13GSdoypW0v6YG0z18lfaRZXk2zFtC+0hWwtkFSe7L7uj+QkvoAu0bErBQYX4+IPSV1AP4maSzwSWAn4GNAN+B54IZ6+W4J/Abom/LqEhGLJF0LvBkRP0vb3QxcFhGPStqa7Fe2HwUuAB6NiIskHQqsEqQb8Y1UxobAJEl/jIjXgI2BqRFxhqTzU96nkj3g+aSIeEHSXsDVwOfKeBnNWpyDuq3OhpKeSvN/Ba4n6xaZGBGzUvoXgI/X9ZcDmwK9gb7ALRFRC8yV9HAD+e8NTKjLKyIau0/4AcDO0sqG+CaSOqUyvpz2vVfS4hKOaYikL6X5XqmurwErgFtT+k3AHZI6puO9LVd2hxLKMKsIB3VbnWURsVs+IQW3t/JJwGkR8WC97Q5h9bcKVgnbQNZVuE9ELGugLiXf60JSP7IPiH0i4m1J44ENGtk8UrlL6r8GZtXKferWHB4ETpa0HoCkHSVtDEwABqQ+9+7A/zSw72PAZyVtl/btktKXAp1y240l6wohbbdbmp0AHJfSDgY6r6aumwKLU0D/CNk3hTo1QN23ja+Sdeu8AcySdFQqQ5I+sZoyzCrGQd2aw3Vk/eVT08OQf032LfBO4AXgWeAa4C/1d4yIBWT94HdIepr3uz/uAb5Ud6EUGALskS7EPs/7o3B+CPSVNJWsG+iV1dT1AaC9pGeAi4HHc+veAnaRNIWsz/yilH4ccGKq3zT8SEGrYr5Lo5lZgbilbmZWIA7qZmYF4qBuZlYgDupmZgXioG5mViAO6mZmBeKgbmZWIP8PnM/OWzF2Lv8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_lr = LogisticRegression(random_state=42)\n",
    "model_lr.fit(x_train_wiki, y_train)\n",
    "predictions = model_lr.predict(x_test_wiki)\n",
    "evaluate(y_test, predictions )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9714f0c7-3bf2-454f-9868-2b7aef088a43",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.8908    0.9017    0.8962      1393\n",
      "         1.0     0.9004    0.8894    0.8949      1393\n",
      "\n",
      "    accuracy                         0.8955      2786\n",
      "   macro avg     0.8956    0.8955    0.8955      2786\n",
      "weighted avg     0.8956    0.8955    0.8955      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53492a87-a2c5-455d-8d2b-ffcd1733ed92",
   "metadata": {},
   "source": [
    "#### Using **SVM** model with the **default hyperparameters**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4de73deb-14ab-44bb-82e0-71c3212f2002",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_svm =svm.SVC(random_state=42)\n",
    "model_svm.fit(x_train_wiki, y_train)\n",
    "predictions = model_svm.predict(x_test_wiki)\n",
    "evaluate(y_test, predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "df77f6bb-6096-4e1b-933e-9fa97d5307fd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.9014    0.9253    0.9132      1393\n",
      "         1.0     0.9233    0.8988    0.9109      1393\n",
      "\n",
      "    accuracy                         0.9121      2786\n",
      "   macro avg     0.9124    0.9121    0.9120      2786\n",
      "weighted avg     0.9124    0.9121    0.9120      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "7d2f18b1-d0d8-4dc1-afc1-ea40d1911ad4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best C: 2\n",
      "Best l1_ratio: none\n",
      "Best solver: lbfgs\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Hyperparemeter tuning for Logistic Regression model\n",
    "grid_lr_wiki = {\"C\":[1, 2, 3, 4], 'l1_ratio':['none', 0, 0.1], 'penalty':['l2', 'elasticnet'],\n",
    "       'solver':['lbfgs', 'saga']}\n",
    "\n",
    "model_lr_1000 = LogisticRegression(max_iter=1000, random_state=42)\n",
    "\n",
    "\n",
    "logreg_cv_glove_wiki=GridSearchCV(model_lr_1000, grid_lr_wiki, n_jobs=-1)\n",
    "logreg_cv_glove_wiki.fit(x_train_wiki, y_train)\n",
    "\n",
    "print('Best C:', logreg_cv_glove_wiki.best_estimator_.get_params()['C'])\n",
    "print('Best l1_ratio:', logreg_cv_glove_wiki.best_estimator_.get_params()['l1_ratio'])\n",
    "print('Best solver:', logreg_cv_glove_wiki.best_estimator_.get_params()['solver'])\n",
    "\n",
    "predictions = logreg_cv_glove_wiki.predict(x_test_wiki)\n",
    "evaluate(y_test, predictions )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "e7d17768-de6f-4b89-952a-0755a83b881f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.8914    0.9017    0.8965      1393\n",
      "         1.0     0.9005    0.8902    0.8953      1393\n",
      "\n",
      "    accuracy                         0.8959      2786\n",
      "   macro avg     0.8960    0.8959    0.8959      2786\n",
      "weighted avg     0.8960    0.8959    0.8959      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6f836e8d-3993-4ff9-a3da-dceec6a99541",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best KERNEL: rbf\n",
      "Best C: 35\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Hyperparameter tuning for SVM model\n",
    "\n",
    "grid_svm = [\n",
    "    {'kernel':('linear', 'rbf')},\n",
    "    {'C':[1, 5, 11, 15, 19, 23, 27, 31, 35, 39, 43]},\n",
    "]    \n",
    "\n",
    "model_svm = svm.SVC(random_state=42)\n",
    "\n",
    "\n",
    "svm_model_cv_wiki = GridSearchCV(model_svm, grid_svm, n_jobs=-1)\n",
    "\n",
    "svm_model_cv_wiki.fit(x_train_wiki, y_train)\n",
    "\n",
    "print('Best KERNEL:', svm_model_cv_wiki.best_estimator_.get_params()['kernel'])\n",
    "print('Best C:', svm_model_cv_wiki.best_estimator_.get_params()['C'])\n",
    "\n",
    "predictions = svm_model_cv_wiki.predict(x_test_wiki)\n",
    "evaluate(y_test, predictions )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a7d98688-3d61-43cc-8bbe-6d41141366e4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0     0.9374    0.9777    0.9571      1393\n",
      "         1.0     0.9767    0.9347    0.9552      1393\n",
      "\n",
      "    accuracy                         0.9562      2786\n",
      "   macro avg     0.9571    0.9562    0.9562      2786\n",
      "weighted avg     0.9571    0.9562    0.9562      2786\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, predictions, digits=4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43156764-f7f8-4e00-8e17-5840feccbf3d",
   "metadata": {},
   "source": [
    "### **Conclusions**: \n",
    "#### - **All F1 scores** have **high values (above 89.0%)**.\n",
    "#### - **All tuned** models have **higher values** (some only slightly higher) for **F1 score** than the **corresponding non-tuned** models, as **expected**.\n",
    "#### - **Comparing** the used metrics regarding pre-trained Glove's word embeddings based on **wikipedia** and on **twitter** datasets, in **this context**, **there aren't noticeable differences**.\n",
    "#### - The **best F1 scores** were obtained using the **SVM** model with **hyperparameter tuning** (slightly above 95%)."
   ]
  }
 ],
 "metadata": {
  "environment": {
   "kernel": "python3",
   "name": "tf2-gpu.2-6.m91",
   "type": "gcloud",
   "uri": "gcr.io/deeplearning-platform-release/tf2-gpu.2-6:m91"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}