movie_sentiment.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Análisis de sentimientos de películas. \n",
    "Se tiene comentarios de películas\n",
    "en español, que serán clasificadas en las categorías de positivo, negativo. "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Instalación e importación de librerías."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: inflect in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (6.0.2)\n",
      "Requirement already satisfied: pydantic>=1.9.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from inflect) (1.10.7)\n",
      "Requirement already satisfied: typing-extensions>=4.2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from pydantic>=1.9.1->inflect) (4.2.0)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: ydata-profiling in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (4.1.2)\n",
      "Requirement already satisfied: PyYAML<6.1,>=5.0.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (6.0)\n",
      "Requirement already satisfied: pydantic<1.11,>=1.8.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (1.10.7)\n",
      "Requirement already satisfied: numpy<1.24,>=1.16.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (1.22.3)\n",
      "Collecting visions[type_image_path]==0.7.5\n",
      "  Using cached visions-0.7.5-py3-none-any.whl (102 kB)\n",
      "Requirement already satisfied: requests<2.29,>=2.24.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (2.27.1)\n",
      "Requirement already satisfied: imagehash==4.3.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (4.3.1)\n",
      "Requirement already satisfied: pandas!=1.4.0,<1.6,>1.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (1.5.3)\n",
      "Requirement already satisfied: htmlmin==0.1.12 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (0.1.12)\n",
      "Requirement already satisfied: matplotlib<3.7,>=3.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (3.5.1)\n",
      "Requirement already satisfied: phik<0.13,>=0.11.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (0.12.3)\n",
      "Requirement already satisfied: statsmodels<0.14,>=0.13.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (0.13.2)\n",
      "Requirement already satisfied: multimethod<1.10,>=1.4 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (1.4)\n",
      "Requirement already satisfied: jinja2<3.2,>=2.11.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (3.0.1)\n",
      "Requirement already satisfied: seaborn<0.13,>=0.10.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (0.12.2)\n",
      "Requirement already satisfied: typeguard<2.14,>=2.13.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (2.13.3)\n",
      "Requirement already satisfied: scipy<1.10,>=1.4.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (1.8.0)\n",
      "Requirement already satisfied: tqdm<4.65,>=4.48.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from ydata-profiling) (4.64.0)\n",
      "Requirement already satisfied: PyWavelets in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from imagehash==4.3.1->ydata-profiling) (1.4.1)\n",
      "Requirement already satisfied: pillow in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from imagehash==4.3.1->ydata-profiling) (9.1.0)\n",
      "Requirement already satisfied: attrs>=19.3.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from visions[type_image_path]==0.7.5->ydata-profiling) (21.4.0)\n",
      "Requirement already satisfied: networkx>=2.4 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from visions[type_image_path]==0.7.5->ydata-profiling) (2.8)\n",
      "Requirement already satisfied: tangled-up-in-unicode>=0.0.4 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from visions[type_image_path]==0.7.5->ydata-profiling) (0.2.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from jinja2<3.2,>=2.11.1->ydata-profiling) (2.1.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (4.33.3)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (0.11.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (2.8.2)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (3.0.8)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (1.4.2)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib<3.7,>=3.2->ydata-profiling) (21.3)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from pandas!=1.4.0,<1.6,>1.1->ydata-profiling) (2021.1)\n",
      "Requirement already satisfied: joblib>=0.14.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from phik<0.13,>=0.11.1->ydata-profiling) (1.2.0)\n",
      "Requirement already satisfied: typing-extensions>=4.2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from pydantic<1.11,>=1.8.1->ydata-profiling) (4.2.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<2.29,>=2.24.0->ydata-profiling) (1.26.9)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<2.29,>=2.24.0->ydata-profiling) (2.10)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<2.29,>=2.24.0->ydata-profiling) (2021.10.8)\n",
      "Requirement already satisfied: charset-normalizer~=2.0.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<2.29,>=2.24.0->ydata-profiling) (2.0.12)\n",
      "Requirement already satisfied: patsy>=0.5.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from statsmodels<0.14,>=0.13.2->ydata-profiling) (0.5.2)\n",
      "Requirement already satisfied: colorama in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from tqdm<4.65,>=4.48.2->ydata-profiling) (0.4.6)\n",
      "Requirement already satisfied: six in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from patsy>=0.5.2->statsmodels<0.14,>=0.13.2->ydata-profiling) (1.16.0)\n",
      "Installing collected packages: visions\n",
      "  Attempting uninstall: visions\n",
      "    Found existing installation: visions 0.4.1\n",
      "    Uninstalling visions-0.4.1:\n",
      "      Successfully uninstalled visions-0.4.1\n",
      "Successfully installed visions-0.7.5\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "pandas-profiling 2.7.1 requires visions[type_image_path]==0.4.1, but you have visions 0.7.5 which is incompatible.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: contractions in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (0.1.73)\n",
      "Requirement already satisfied: textsearch>=0.0.21 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from contractions) (0.0.24)\n",
      "Requirement already satisfied: anyascii in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from textsearch>=0.0.21->contractions) (0.3.2)\n",
      "Requirement already satisfied: pyahocorasick in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from textsearch>=0.0.21->contractions) (2.0.0)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: nltk in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (3.7)\n",
      "Requirement already satisfied: regex>=2021.8.3 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from nltk) (2022.7.9)\n",
      "Requirement already satisfied: tqdm in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from nltk) (4.64.0)\n",
      "Requirement already satisfied: joblib in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from nltk) (1.2.0)\n",
      "Requirement already satisfied: click in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from nltk) (8.0.1)\n",
      "Requirement already satisfied: colorama in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from click->nltk) (0.4.6)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Collecting scikit-plot\n",
      "  Downloading scikit_plot-0.3.7-py3-none-any.whl (33 kB)\n",
      "Requirement already satisfied: scikit-learn>=0.18 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-plot) (1.0.2)\n",
      "Requirement already satisfied: joblib>=0.10 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-plot) (1.2.0)\n",
      "Requirement already satisfied: matplotlib>=1.4.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-plot) (3.5.1)\n",
      "Requirement already satisfied: scipy>=0.9 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-plot) (1.8.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (2.8.2)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (0.11.0)\n",
      "Requirement already satisfied: numpy>=1.17 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (1.22.3)\n",
      "Requirement already satisfied: pillow>=6.2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (9.1.0)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (3.0.8)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (1.4.2)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (21.3)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from matplotlib>=1.4.0->scikit-plot) (4.33.3)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from scikit-learn>=0.18->scikit-plot) (3.1.0)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from python-dateutil>=2.7->matplotlib>=1.4.0->scikit-plot) (1.16.0)\n",
      "Installing collected packages: scikit-plot\n",
      "Successfully installed scikit-plot-0.3.7\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: spacy in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (3.5.1)\n",
      "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (3.0.8)\n",
      "Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (1.1.1)\n",
      "Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (3.0.12)\n",
      "Requirement already satisfied: srsly<3.0.0,>=2.4.3 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (2.4.6)\n",
      "Requirement already satisfied: pathy>=0.10.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (0.10.1)\n",
      "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (2.0.7)\n",
      "Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (2.0.8)\n",
      "Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (1.0.4)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (21.3)\n",
      "Requirement already satisfied: tqdm<5.0.0,>=4.38.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (4.64.0)\n",
      "Requirement already satisfied: jinja2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (3.0.1)\n",
      "Requirement already satisfied: pydantic!=1.8,!=1.8.1,<1.11.0,>=1.7.4 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (1.10.7)\n",
      "Requirement already satisfied: typer<0.8.0,>=0.3.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (0.7.0)\n",
      "Requirement already satisfied: thinc<8.2.0,>=8.1.8 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (8.1.9)\n",
      "Requirement already satisfied: requests<3.0.0,>=2.13.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (2.27.1)\n",
      "Requirement already satisfied: numpy>=1.15.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (1.22.3)\n",
      "Requirement already satisfied: setuptools in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (62.1.0)\n",
      "Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (3.3.0)\n",
      "Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (6.3.0)\n",
      "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from spacy) (1.0.9)\n",
      "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from packaging>=20.0->spacy) (3.0.8)\n",
      "Requirement already satisfied: typing-extensions>=4.2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from pydantic!=1.8,!=1.8.1,<1.11.0,>=1.7.4->spacy) (4.2.0)\n",
      "Requirement already satisfied: charset-normalizer~=2.0.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<3.0.0,>=2.13.0->spacy) (2.0.12)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<3.0.0,>=2.13.0->spacy) (2021.10.8)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<3.0.0,>=2.13.0->spacy) (1.26.9)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from requests<3.0.0,>=2.13.0->spacy) (2.10)\n",
      "Requirement already satisfied: confection<1.0.0,>=0.0.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from thinc<8.2.0,>=8.1.8->spacy) (0.0.4)\n",
      "Requirement already satisfied: blis<0.8.0,>=0.7.8 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from thinc<8.2.0,>=8.1.8->spacy) (0.7.9)\n",
      "Requirement already satisfied: colorama in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from tqdm<5.0.0,>=4.38.0->spacy) (0.4.6)\n",
      "Requirement already satisfied: click<9.0.0,>=7.1.1 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from typer<0.8.0,>=0.3.0->spacy) (8.0.1)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from jinja2->spacy) (2.1.1)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install inflect\n",
    "%pip install ydata-profiling\n",
    "%pip install contractions\n",
    "%pip install nltk\n",
    "%pip install scikit-plot\n",
    "%pip install spacy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     C:\\Users\\fesar\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\fesar\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package wordnet to\n",
      "[nltk_data]     C:\\Users\\fesar\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sys\n",
    "from ydata_profiling import ProfileReport\n",
    "\n",
    "import re, string, unicodedata\n",
    "import contractions\n",
    "from nltk import word_tokenize, sent_tokenize\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import LancasterStemmer, WordNetLemmatizer\n",
    "\n",
    "from sklearn.model_selection import train_test_split,GridSearchCV\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer, HashingVectorizer\n",
    "from sklearn.pipeline import Pipeline, FeatureUnion\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import BaggingClassifier, RandomForestClassifier, AdaBoostClassifier\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "from sklearn.base import BaseEstimator, ClassifierMixin\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import string\n",
    "\n",
    "\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn.metrics import ConfusionMatrixDisplay, precision_score, recall_score, f1_score\n",
    "\n",
    "import nltk\n",
    "import nltk.data\n",
    "import spacy\n",
    "# Punkt permite separar un texto en frases.\n",
    "nltk.download('punkt')\n",
    "nltk.download('stopwords')\n",
    "nltk.download('wordnet')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Perfilamiento y entendimiento de los datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    " # Uso de la libreria pandas para la lectura de archivos\n",
    "data=pd.read_csv('data/MovieReviews.csv', sep=',', encoding = 'utf-8')\n",
    "# Asignación a una nueva variable de los datos leidos\n",
    "data_t=data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "textos = data_t.copy()\n",
    "textos['Conteo'] = [len(x) for x in textos['review_es']]\n",
    "textos['Moda'] = [max(set(x), key=x.count) for x in textos['review_es']]\n",
    "textos['Max'] = [[max([len(x) for x in i.split(' ')])][0] for i in textos['review_es']]\n",
    "textos['Min'] = [[min([len(x) for x in i.split(' ')])][0] for i in textos['review_es']]\n",
    "\n",
    "# Se realiza un perfilamiento de los datos con la librería pandas profiling\n",
    "#ProfileReport(textos)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Preparación de datos"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para poder realizar el pre-procesamiento de los datos, se pasará por tres etapas:\n",
    "- Limpieza de los datos.\n",
    "- Tokenización.\n",
    "- Normalización."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.1 Limpieza de los datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>review_es</th>\n",
       "      <th>sentimiento</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Si está buscando una película de guerra típica...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Supongo que algunos directores de películas de...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es difícil contarle más sobre esta película si...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>La película comienza muy lentamente, con el es...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta película es verdadera acción en su máxima...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           review_es sentimiento\n",
       "0  Si está buscando una película de guerra típica...    positivo\n",
       "1  Supongo que algunos directores de películas de...    positivo\n",
       "2  Es difícil contarle más sobre esta película si...    positivo\n",
       "3  La película comienza muy lentamente, con el es...    positivo\n",
       "4  Esta película es verdadera acción en su máxima...    positivo"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Eliminar primera columna sin nombre que es el id\n",
    "data_t = data_t.drop(data_t.columns[0], axis=1)\n",
    "data_t.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_words = set(stopwords.words('spanish'))\n",
    "def remove_non_ascii(words):\n",
    "    \"\"\"Remove non-ASCII characters from list of tokenized words\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = word.encode('utf-8', 'ignore').decode('utf-8', 'ignore')\n",
    "        new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def to_lowercase(words):\n",
    "    \"\"\"Convert all characters to lowercase from list of tokenized words\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = word.lower()\n",
    "        new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def remove_punctuation(words):\n",
    "    \"\"\"Remove punctuation from list of tokenized words\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = re.sub(r'[^\\w\\s]', '', word)\n",
    "        if new_word != '':\n",
    "            new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def replace_numbers(words):\n",
    "    \"\"\"Replace all interger occurrences in list of tokenized words with textual representation\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = re.sub(r'\\d+', '', word)\n",
    "        if new_word != '':\n",
    "            new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def remove_stopwords(words):\n",
    "    \"\"\"Remove stop words from list of tokenized words\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        if word not in stop_words:\n",
    "            new_words.append(word)\n",
    "    return new_words\n",
    "\n",
    "def preprocessing(words):\n",
    "    words = to_lowercase(words)\n",
    "    words = replace_numbers(words)\n",
    "    words = remove_punctuation(words)\n",
    "    words = remove_non_ascii(words)\n",
    "    words = remove_stopwords(words)\n",
    "    return words"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.2 Tokenización"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>review_es</th>\n",
       "      <th>sentimiento</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Si está buscando una película de guerra típica...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Supongo que algunos directores de películas de...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es difícil contarle más sobre esta película si...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>La película comienza muy lentamente, con el es...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta película es verdadera acción en su máxima...</td>\n",
       "      <td>positivo</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           review_es sentimiento\n",
       "0  Si está buscando una película de guerra típica...    positivo\n",
       "1  Supongo que algunos directores de películas de...    positivo\n",
       "2  Es difícil contarle más sobre esta película si...    positivo\n",
       "3  La película comienza muy lentamente, con el es...    positivo\n",
       "4  Esta película es verdadera acción en su máxima...    positivo"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_t.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>review_es</th>\n",
       "      <th>sentimiento</th>\n",
       "      <th>words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Si está buscando una película de guerra típica...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[si, buscando, película, guerra, típica, así, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Supongo que algunos directores de películas de...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[supongo, directores, películas, lujo, sentado...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es difícil contarle más sobre esta película si...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[difícil, contarle, película, estropearlalo, d...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>La película comienza muy lentamente, con el es...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[película, comienza, lentamente, estilo, vida,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta película es verdadera acción en su máxima...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[película, verdadera, acción, máxima, expresió...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           review_es sentimiento  \\\n",
       "0  Si está buscando una película de guerra típica...    positivo   \n",
       "1  Supongo que algunos directores de películas de...    positivo   \n",
       "2  Es difícil contarle más sobre esta película si...    positivo   \n",
       "3  La película comienza muy lentamente, con el es...    positivo   \n",
       "4  Esta película es verdadera acción en su máxima...    positivo   \n",
       "\n",
       "                                               words  \n",
       "0  [si, buscando, película, guerra, típica, así, ...  \n",
       "1  [supongo, directores, películas, lujo, sentado...  \n",
       "2  [difícil, contarle, película, estropearlalo, d...  \n",
       "3  [película, comienza, lentamente, estilo, vida,...  \n",
       "4  [película, verdadera, acción, máxima, expresió...  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_t['review_es'] = data_t['review_es'].apply(contractions.fix) #Aplica la corrección de las contracciones  \n",
    "data_t['words'] = data_t['review_es'].apply(word_tokenize)\n",
    "data_t['words'] = data_t['words'].apply(preprocessing)\n",
    "data_t.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  3.3. Normalización"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>review_es</th>\n",
       "      <th>sentimiento</th>\n",
       "      <th>words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Si está buscando una película de guerra típica...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[si, buscando, películ, guerr, típica, así, no...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Supongo que algunos directores de películas de...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[supongo, direct, película, lujo, sentado, bus...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es difícil contarle más sobre esta película si...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[difícil, contarl, películ, estropearlalo, dis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>La película comienza muy lentamente, con el es...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[películ, comienz, lenta, estilo, vid, wallac,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta película es verdadera acción en su máxima...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>[películ, verdader, acción, máxima, expresión,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           review_es sentimiento  \\\n",
       "0  Si está buscando una película de guerra típica...    positivo   \n",
       "1  Supongo que algunos directores de películas de...    positivo   \n",
       "2  Es difícil contarle más sobre esta película si...    positivo   \n",
       "3  La película comienza muy lentamente, con el es...    positivo   \n",
       "4  Esta película es verdadera acción en su máxima...    positivo   \n",
       "\n",
       "                                               words  \n",
       "0  [si, buscando, películ, guerr, típica, así, no...  \n",
       "1  [supongo, direct, película, lujo, sentado, bus...  \n",
       "2  [difícil, contarl, películ, estropearlalo, dis...  \n",
       "3  [películ, comienz, lenta, estilo, vid, wallac,...  \n",
       "4  [películ, verdader, acción, máxima, expresión,...  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def stem_words(words):\n",
    "    \"\"\"Stem words in list of tokenized words\"\"\"\n",
    "    stemmer = LancasterStemmer()\n",
    "    stems = []\n",
    "    for word in words:\n",
    "        stem = stemmer.stem(word)\n",
    "        stems.append(stem)\n",
    "    return stems\n",
    "\n",
    "def lemmatize_verbs(words):\n",
    "    \"\"\"Lemmatize verbs in list of tokenized words\"\"\"\n",
    "    lemmatizer = WordNetLemmatizer()\n",
    "    lemmas = []\n",
    "    for word in words:\n",
    "        lemma = lemmatizer.lemmatize(word, pos='v')\n",
    "        lemmas.append(lemma)\n",
    "    return lemmas\n",
    "\n",
    "def stem_and_lemmatize(words):\n",
    "    stems = stem_words(words)\n",
    "    lemmas = lemmatize_verbs(words)\n",
    "    return stems + lemmas\n",
    "\n",
    "data_t['words'] = data_t['words'].apply(stem_and_lemmatize) #Aplica lematización y Eliminación de Prefijos y Sufijos.\n",
    "data_t.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  3.4 Selección de campos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>review_es</th>\n",
       "      <th>sentimiento</th>\n",
       "      <th>words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Si está buscando una película de guerra típica...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>si buscando películ guerr típica así not afici...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Supongo que algunos directores de películas de...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>supongo direct película lujo sentado buscando ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es difícil contarle más sobre esta película si...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>difícil contarl películ estropearlalo disfruté...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>La película comienza muy lentamente, con el es...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>películ comienz lenta estilo vid wallac napalm...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta película es verdadera acción en su máxima...</td>\n",
       "      <td>positivo</td>\n",
       "      <td>películ verdader acción máxima expresión mej e...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>\"Criminal decente ordinario\" es triste porque ...</td>\n",
       "      <td>negativo</td>\n",
       "      <td>crimin dec ordinario trist obvia tratando triu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>Savage Island (2003) es una película coja.Es m...</td>\n",
       "      <td>negativo</td>\n",
       "      <td>sav island películ coja video video cas ilumin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>Quien escribió el script para esta película no...</td>\n",
       "      <td>negativo</td>\n",
       "      <td>escribió script películ merec trabaj hollywood...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>in a TV-movie 70's kind of way It is one of th...</td>\n",
       "      <td>negativo</td>\n",
       "      <td>in tvmovie s kind of way it is on of thos movy...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>Película decepcionante, predecible en la que u...</td>\n",
       "      <td>negativo</td>\n",
       "      <td>películ decepc predec muj mc teer viaj hij pue...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              review_es sentimiento  \\\n",
       "0     Si está buscando una película de guerra típica...    positivo   \n",
       "1     Supongo que algunos directores de películas de...    positivo   \n",
       "2     Es difícil contarle más sobre esta película si...    positivo   \n",
       "3     La película comienza muy lentamente, con el es...    positivo   \n",
       "4     Esta película es verdadera acción en su máxima...    positivo   \n",
       "...                                                 ...         ...   \n",
       "4995  \"Criminal decente ordinario\" es triste porque ...    negativo   \n",
       "4996  Savage Island (2003) es una película coja.Es m...    negativo   \n",
       "4997  Quien escribió el script para esta película no...    negativo   \n",
       "4998  in a TV-movie 70's kind of way It is one of th...    negativo   \n",
       "4999  Película decepcionante, predecible en la que u...    negativo   \n",
       "\n",
       "                                                  words  \n",
       "0     si buscando películ guerr típica así not afici...  \n",
       "1     supongo direct película lujo sentado buscando ...  \n",
       "2     difícil contarl películ estropearlalo disfruté...  \n",
       "3     películ comienz lenta estilo vid wallac napalm...  \n",
       "4     películ verdader acción máxima expresión mej e...  \n",
       "...                                                 ...  \n",
       "4995  crimin dec ordinario trist obvia tratando triu...  \n",
       "4996  sav island películ coja video video cas ilumin...  \n",
       "4997  escribió script películ merec trabaj hollywood...  \n",
       "4998  in tvmovie s kind of way it is on of thos movy...  \n",
       "4999  películ decepc predec muj mc teer viaj hij pue...  \n",
       "\n",
       "[5000 rows x 3 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_t['words'] = data_t['words'].apply(lambda x: ' '.join(map(str, x)))\n",
    "data_t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1       1\n",
       "2       1\n",
       "3       1\n",
       "4       1\n",
       "       ..\n",
       "4995    0\n",
       "4996    0\n",
       "4997    0\n",
       "4998    0\n",
       "4999    0\n",
       "Name: sentimiento, Length: 5000, dtype: int32"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_data, y_data = data_t['words'],data_t['sentimiento']\n",
    "y_data = (y_data == 'positivo').astype(int)\n",
    "y_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5000, 90534)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummy = CountVectorizer(binary=True)\n",
    "X_dummy = dummy.fit_transform(X_data)\n",
    "print(X_dummy.shape)\n",
    "X_dummy.toarray()[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5000, 90534)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count = CountVectorizer()\n",
    "X_count = count.fit_transform(X_data)\n",
    "print(X_count.shape)\n",
    "X_count.toarray()[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5000, 90534)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., ..., 0., 0., 0.])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf_idf = TfidfVectorizer()\n",
    "X_tf_idf = tf_idf.fit_transform(X_data)\n",
    "\n",
    "print(X_tf_idf.shape)\n",
    "X_tf_idf.toarray()[0]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Creación y Aplicación del modelo"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.1 Entrenando y evaluando un modelo usando Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_words = set(stopwords.words('spanish'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "positivo    0.5\n",
       "negativo    0.5\n",
       "Name: sentimiento, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_t['sentimiento'].value_counts(dropna = False, normalize = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size = 0.2, stratify = y_data, random_state = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4000,)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.5\n",
       "1    0.5\n",
       "Name: sentimiento, dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(y_train).value_counts(normalize = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipeline = Pipeline(steps = [\n",
    "    ('vectorizer', CountVectorizer(tokenizer = word_tokenize, stop_words = stop_words, lowercase = True)),\n",
    "    ('classifier', RandomForestClassifier(random_state = 4))\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {\n",
    "    'vectorizer': [CountVectorizer(tokenizer = word_tokenize, stop_words = stop_words), TfidfVectorizer(tokenizer = word_tokenize, stop_words = stop_words)],\n",
    "    'vectorizer__lowercase': [True, False],\n",
    "    'classifier__n_estimators': [50, 100, 150, 200, 250, 300],\n",
    "    'classifier__criterion': ['gini', 'entropy'],\n",
    "    'classifier__max_depth': [20, 40, 60, 80]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "search = GridSearchCV(pipeline, param_grid, scoring = ['precision', 'recall', 'f1'], refit = 'f1', cv = 10, return_train_score = True, verbose = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 10 folds for each of 192 candidates, totalling 1920 fits\n"
     ]
    }
   ],
   "source": [
    "search.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test_search_predict = search.best_estimator_.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABPsAAAMtCAYAAADpJHSLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAACCPUlEQVR4nOz9ebhWVf0//j9vpgMHOKCIggoiCggKCuKAZJBaoEjOUyRSipaSoqHG2xxAE8x5yDJNwaHUnEtzFgc0pwIhCRFF+BTveDudE1LIcH5/+PP+emQQFDx483hc176us/dee63Xvs9/z2utvQrV1dXVAQAAAAC+8urUdgEAAAAAwJoh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBJRr7YLYPmWLl2af/7zn2natGkKhUJtlwMAAABALaqurs6///3vbLrppqlTZ8Xz94R966h//vOfadOmTW2XAQAAAMA6ZM6cOdl8881XeF/Yt45q2rRpko/+gRUVFbVcDQAAAAC1qaqqKm3atClmRisi7FtHfbx0t6KiQtgHAAAAQJJ85ufebNABAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACWiXm0XwMptd/ZDqVNWXttlAAAAAKzzZo0dUNsl1Doz+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRKyRsK9v374ZPnz4mujqC43brl27XHbZZcXzQqGQe+655wuPs6b6AQAAAIC16Su9G+9dd92V+vXrr/D+3Llzs8EGG6xyf+ecc07uueeeTJo06Qv1AwAAAAC14Ssd9m244YYrvd+qVas1Ms6a6gcAAAAA1qY19s2+pUuX5rTTTsuGG26YVq1a5ZxzzkmSzJo1K4VCocZsuffffz+FQiETJkxIkkyYMCGFQiEPPfRQunfvnkaNGmWPPfbIvHnz8qc//SmdO3dORUVFvvOd72TBggXFfj5r+fCnl9+efvrp6dixY8rLy9O+ffuceeaZWbRoUZJk3LhxGTVqVCZPnpxCoZBCoZBx48Ytt58pU6Zkjz32SKNGjdKiRYsce+yxmT9/fvH+kCFDsv/+++eiiy5K69at06JFi5xwwgnFsQAAAABgbVhjM/vGjx+fU045Jc8//3yee+65DBkyJL17906HDh1WuY9zzjknV111VcrLy3PooYfm0EMPTVlZWX77299m/vz5OeCAA3LllVfm9NNP/1w1Nm3aNOPGjcumm26aKVOmZOjQoWnatGlOO+20HHbYYZk6dWoefPDBPProo0mSZs2aLdPHBx98kH79+qVXr1558cUXM2/evBxzzDEZNmxYMRxMkieeeCKtW7fOE088kddffz2HHXZYdthhhwwdOnS5tS1cuDALFy4snldVVX2udwQAAABg/bXGZvZ169YtZ599djp06JDBgwenZ8+eeeyxx1arj/POOy+9e/dO9+7dc/TRR+fJJ5/ML3/5y3Tv3j277757Dj744DzxxBOfu8af/vSn2W233dKuXbsMHDgwI0aMyO23354kadSoUZo0aZJ69eqlVatWadWqVRo1arRMH7/97W/z3//+NzfeeGO222677LHHHrnqqqty00035V//+lex3QYbbJCrrroq22yzTfbdd98MGDBgpb/HmDFj0qxZs+LRpk2bz/2eAAAAAKyf1mjY90mtW7fOvHnzPncfm2yySXG57SevrW6fn3Tbbbeld+/eadWqVZo0aZKf/vSnmT179mr1MW3atGy//fZp3Lhx8Vrv3r2zdOnSTJ8+vXht2223Td26dYvnn/V7jBw5MpWVlcVjzpw5q1UXAAAAAKyxsO/Tu+IWCoUsXbo0dep8NER1dXXx3oq+XffJPgqFwgr7/Dyee+65DBo0KPvss0/++Mc/5q9//WvOOOOMfPjhh5+rv8+yurWXlZWloqKixgEAAAAAq2ONhX0r0rJlyyTJ3Llzi9c+uVnHl+XZZ5/NFltskTPOOCM9e/ZMhw4d8tZbb9Vo06BBgyxZsmSl/XTu3DmTJ0/OBx98ULw2ceLE1KlTJ506dVortQMAAADAqljrYV+jRo2y6667ZuzYsZk2bVqefPLJ/PSnP13bwy6jQ4cOmT17dm699dbMnDkzV1xxRe6+++4abdq1a5c333wzkyZNyttvv11jw4yPDRo0KA0bNsxRRx2VqVOn5oknnsiPfvSjHHnkkdlkk02+rNcBAAAAgGWs9bAvSa6//vosXrw4O+64Y4YPH57zzjvvyxi2hm9/+9s5+eSTM2zYsOywww559tlnc+aZZ9Zoc9BBB6V///75xje+kZYtW+Z3v/vdMv2Ul5fnoYceyrvvvpuddtopBx98cPbcc89cddVVX9arAAAAAMByFao/+TE91hlVVVUf7co7/PbUKSuv7XIAAAAA1nmzxg6o7RLWmo+zosrKypXu9fClzOwDAAAAANY+YR8AAAAAlIh6tV0AKzd1VL+VTs0EAAAAgI+Z2QcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACWiXm0XwMptd/ZDqVNWXttlAAAAwDJmjR1Q2yUAn2JmHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUiFoL+/r27Zvhw4ev8H67du1y2WWXfWn1fBFfpVoBAAAAKF3r7G68L774Yho3blzbZdQwbty4DB8+PO+//36N6+tirQAAAACsf9bZsK9ly5a1XcIq+yrVCgAAAEDpqtVv9i1evDjDhg1Ls2bNstFGG+XMM89MdXV1kmWXxr7//vs57rjjsskmm6Rhw4bZbrvt8sc//rF4/5lnnsnuu++eRo0apU2bNjnxxBPzwQcfFO+3a9cu559/fr7//e+nadOmadu2bX79618X78+aNSuFQiF33XVXvvGNb6S8vDzbb799nnvuuSTJhAkT8r3vfS+VlZUpFAopFAo555xzlltroVDIddddlwMOOCDl5eXp0KFD7rvvvrXwCwIAAADA/6dWw77x48enXr16eeGFF3L55ZfnkksuyXXXXbdMu6VLl2bvvffOxIkTc/PNN+fVV1/N2LFjU7du3STJzJkz079//xx00EF55ZVXctttt+WZZ57JsGHDavRz8cUXp2fPnvnrX/+a448/Pj/84Q8zffr0Gm3OOOOMjBgxIpMmTUrHjh1zxBFHZPHixdltt91y2WWXpaKiInPnzs3cuXMzYsSIFb7bqFGjcuihh+aVV17JPvvsk0GDBuXdd99dYfuFCxemqqqqxgEAAAAAq6NWl/G2adMml156aQqFQjp16pQpU6bk0ksvzdChQ2u0e/TRR/PCCy9k2rRp6dixY5Kkffv2xftjxozJoEGDiht+dOjQIVdccUX69OmTX/7yl2nYsGGSZJ999snxxx+fJDn99NNz6aWX5oknnkinTp2KfY0YMSIDBgxI8lFgt+222+b111/PNttsk2bNmqVQKKRVq1af+W5DhgzJEUcckSQ5//zzc8UVV+SFF15I//79l9t+zJgxGTVq1Kr8bAAAAACwXLU6s2/XXXdNoVAonvfq1SszZszIkiVLarSbNGlSNt9882LQ92mTJ0/OuHHj0qRJk+LRr1+/LF26NG+++WaxXbdu3Yp/fxzazZs3r0Zfn2zTunXrJFmmzar4ZD+NGzdORUXFSvsZOXJkKisri8ecOXNWe0wAAAAA1m/r7AYdn9SoUaOV3p8/f36OO+64nHjiicvca9u2bfHv+vXr17hXKBSydOnSGtc+2ebjIPLTbVbFqoz1SWVlZSkrK1vtcQAAAADgY7Ua9j3//PM1zv/85z+nQ4cOxW/xfaxbt275f//v/+W1115b7uy+Hj165NVXX83WW2+9Vutt0KDBMrMOAQAAAGBdUavLeGfPnp1TTjkl06dPz+9+97tceeWVOemkk5Zp16dPn3z961/PQQcdlEceeSRvvvlm/vSnP+XBBx9M8tH395599tkMGzYskyZNyowZM3Lvvfcus0HHF9WuXbvMnz8/jz32WN5+++0sWLBgjfYPAAAAAF9ErYZ9gwcPzn/+85/svPPOOeGEE3LSSSfl2GOPXW7bO++8MzvttFOOOOKIdOnSJaeddlpxll23bt3y5JNP5rXXXsvuu++e7t2756yzzsqmm266Ruvdbbfd8oMf/CCHHXZYWrZsmZ///OdrtH8AAAAA+CIK1dXV1bVdBMuqqqpKs2bN0mb47alTVl7b5QAAAMAyZo0dUNslwHrj46yosrIyFRUVK2xXqzP7AAAAAIA1R9gHAAAAACVC2AcAAAAAJaJebRfAyk0d1W+l67ABAAAA4GNm9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQImoV9sFsHLbnf1Q6pSV13YZAAAAX7pZYwfUdgkAXzlm9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2fQ4LFixI165d07Nnz8yePTvbbLNNFi9eXNtlAQAAALCesxvv5/Diiy9mxx13TP/+/fPNb34z++yzT+rV81MCAAAAULskVJ9Dnz590qdPnyTJ4YcfXsvVAAAAAMBHSnIZ7x133JGuXbumUaNGadGiRfbaa6988MEH6du3b4YPH16j7f77758hQ4YUzxcuXJgRI0Zks802S+PGjbPLLrtkwoQJxfvjxo1L8+bN89BDD6Vz585p0qRJ+vfvn7lz5xbbvPjii/nmN7+ZjTbaKM2aNUufPn3yl7/8ZS2/NQAAAADru5IL++bOnZsjjjgi3//+9zNt2rRMmDAhBx54YKqrq1fp+WHDhuW5557LrbfemldeeSWHHHJI+vfvnxkzZhTbLFiwIBdddFFuuummPPXUU5k9e3ZGjBhRvP/vf/87Rx11VJ555pn8+c9/TocOHbLPPvvk3//+9wrHXbhwYaqqqmocAAAAALA6Sm4Z79y5c7N48eIceOCB2WKLLZIkXbt2XaVnZ8+enRtuuCGzZ8/OpptumiQZMWJEHnzwwdxwww05//zzkySLFi3Kr371q2y11VZJPgoIR48eXexnjz32qNHvr3/96zRv3jxPPvlk9t133+WOPWbMmIwaNWr1XhYAAAAAPqHkZvZtv/322XPPPdO1a9cccsghufbaa/Pee++t0rNTpkzJkiVL0rFjxzRp0qR4PPnkk5k5c2axXXl5eTHoS5LWrVtn3rx5xfN//etfGTp0aDp06JBmzZqloqIi8+fPz+zZs1c49siRI1NZWVk85syZ8zneHgAAAID1WcnN7Ktbt24eeeSRPPvss3n44Ydz5ZVX5owzzsjzzz+fOnXqLLOcd9GiRcW/58+fn7p16+bll19O3bp1a7Rr0qRJ8e/69evXuFcoFGr0e9RRR+Wdd97J5Zdfni222CJlZWXp1atXPvzwwxXWXVZWlrKyss/1zgAAAACQlGDYl3wUvvXu3Tu9e/fOWWedlS222CJ33313WrZsWWMjjSVLlmTq1Kn5xje+kSTp3r17lixZknnz5mX33Xf/3ONPnDgxV199dfbZZ58kyZw5c/L2229/sZcCAAAAgM9QcmHf888/n8ceeyzf+ta3svHGG+f555/P//3f/6Vz585p3LhxTjnllNx///3Zaqutcskll+T9998vPtuxY8cMGjQogwcPzsUXX5zu3bvn//7v//LYY4+lW7duGTBgwCrV0KFDh9x0003p2bNnqqqqcuqpp6ZRo0Zr6Y0BAAAA4CMlF/ZVVFTkqaeeymWXXZaqqqpsscUWufjii7P33ntn0aJFmTx5cgYPHpx69erl5JNPLs7q+9gNN9yQ8847Lz/+8Y/zj3/8IxtttFF23XXXFW6ssTy/+c1vcuyxx6ZHjx5p06ZNzj///Bq79QIAAADA2lCo/vRH7FgnVFVVpVmzZmkz/PbUKSuv7XIAAAC+dLPGrtrqKoD1wcdZUWVlZSoqKlbYruR24wUAAACA9ZWwDwAAAABKRMl9s6/UTB3Vb6VTMwEAAADgY2b2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiahX2wWwctud/VDqlJXXdhkAAMB6aNbYAbVdAgCrycw+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgR60zY17dv3wwfPnytjnHPPfdk6623Tt26dTN8+PCMGzcuzZs3L94/55xzssMOOxTPhwwZkv3333+t1gQAAAAAa8o6E/Z9GY477rgcfPDBmTNnTs4999xl7o8YMSKPPfZYLVQGAAAAAF9cvdou4Msyf/78zJs3L/369cumm2663DZNmjRJkyZN1modixYtSv369dfqGAAAAACsn9bZmX33339/mjVrlltuuaW4nPaiiy5K69at06JFi5xwwglZtGhRsf17772XwYMHZ4MNNkh5eXn23nvvzJgxI0kyYcKENG3aNEmyxx57pFAoZMKECcuM+ellvJ/24osvpmXLlrnggguSJA8++GC+9rWvpXnz5mnRokX23XffzJw5s9h+1qxZKRQKue2229KnT580bNgwt9xyyxr4dQAAAABgWetk2Pfb3/42RxxxRG655ZYMGjQoSfLEE09k5syZeeKJJzJ+/PiMGzcu48aNKz4zZMiQvPTSS7nvvvvy3HPPpbq6Ovvss08WLVqU3XbbLdOnT0+S3HnnnZk7d25222231arp8ccfzze/+c387Gc/y+mnn54k+eCDD3LKKafkpZdeymOPPZY6derkgAMOyNKlS2s8+5Of/CQnnXRSpk2bln79+i23/4ULF6aqqqrGAQAAAACrY51bxvuLX/wiZ5xxRv7whz+kT58+xesbbLBBrrrqqtStWzfbbLNNBgwYkMceeyxDhw7NjBkzct9992XixInFEO+WW25JmzZtcs899+SQQw7JxhtvnCTZcMMN06pVq9Wq6e67787gwYNz3XXX5bDDDiteP+igg2q0u/7669OyZcu8+uqr2W677YrXhw8fngMPPHClY4wZMyajRo1arboAAAAA4JPWqZl9d9xxR04++eQ88sgjNYK+JNl2221Tt27d4nnr1q0zb968JMm0adNSr1697LLLLsX7LVq0SKdOnTJt2rQvVNPzzz+fQw45JDfddFONoC9JZsyYkSOOOCLt27dPRUVF2rVrlySZPXt2jXY9e/b8zHFGjhyZysrK4jFnzpwvVDcAAAAA6591Kuzr3r17WrZsmeuvvz7V1dU17n16U4tCobDMctm1Yauttso222yT66+/vsY3ApNk4MCBeffdd3Pttdfm+eefz/PPP58k+fDDD2u0a9y48WeOU1ZWloqKihoHAAAAAKyOdSrs22qrrfLEE0/k3nvvzY9+9KNVfq5z585ZvHhxMWxLknfeeSfTp09Ply5dvlBNG220UR5//PG8/vrrOfTQQ4uB38f9//SnP82ee+6Zzp0757333vtCYwEAAADAF7FOhX1J0rFjxzzxxBO58847M3z48FV6pkOHDtlvv/0ydOjQPPPMM5k8eXK++93vZrPNNst+++33hWvaeOON8/jjj+fvf/97jjjiiCxevDgbbLBBWrRokV//+td5/fXX8/jjj+eUU075wmMBAAAAwOe1zoV9SdKpU6c8/vjj+d3vfpcf//jHq/TMDTfckB133DH77rtvevXqlerq6jzwwAPLLP/9vFq1apXHH388U6ZMyaBBg1JdXZ1bb701L7/8crbbbrucfPLJufDCC9fIWAAAAADweRSqP/1xPNYJVVVVadasWdoMvz11yspruxwAAGA9NGvsgNouAYD/v4+zosrKypXu9bBOzuwDAAAAAFafsA8AAAAASoSwDwAAAABKRL3aLoCVmzqq30rXYQMAAADAx8zsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAElGvtgtg5bY7+6HUKSuv7TIAAID1zKyxA2q7BAA+BzP7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBErDdhX9++fTN8+PC1Oka7du1y2WWXrdUxAAAAAGBF1puwDwAAAABKnbAPAAAAAErEehn2LVy4MCNGjMhmm22Wxo0bZ5dddsmECRNqtHnmmWey++67p1GjRmnTpk1OPPHEfPDBB8X78+bNy8CBA9OoUaNsueWWueWWW2o8X11dnXPOOSdt27ZNWVlZNt1005x44olfxusBAAAAsJ5aL8O+YcOG5bnnnsutt96aV155JYccckj69++fGTNmJElmzpyZ/v3756CDDsorr7yS2267Lc8880yGDRtW7GPIkCGZM2dOnnjiidxxxx25+uqrM2/evOL9O++8M5deemmuueaazJgxI/fcc0+6du26wpoWLlyYqqqqGgcAAAAArI56tV3Al2327Nm54YYbMnv27Gy66aZJkhEjRuTBBx/MDTfckPPPPz9jxozJoEGDiht6dOjQIVdccUX69OmTX/7yl5k9e3b+9Kc/5YUXXshOO+2UJPnNb36Tzp071xinVatW2WuvvVK/fv20bds2O++88wrrGjNmTEaNGrX2XhwAAACAkrfezeybMmVKlixZko4dO6ZJkybF48knn8zMmTOTJJMnT864ceNq3O/Xr1+WLl2aN998M9OmTUu9evWy4447FvvdZptt0rx58+L5IYcckv/85z9p3759hg4dmrvvvjuLFy9eYV0jR45MZWVl8ZgzZ85a+w0AAAAAKE3r3cy++fPnp27dunn55ZdTt27dGveaNGlSbHPcccct9xt7bdu2zWuvvfaZ47Rp0ybTp0/Po48+mkceeSTHH398Lrzwwjz55JOpX7/+Mu3LyspSVlb2Od8KAAAAANbDsK979+5ZsmRJ5s2bl9133325bXr06JFXX301W2+99XLvb7PNNlm8eHFefvnl4jLe6dOn5/3336/RrlGjRhk4cGAGDhyYE044Idtss02mTJmSHj16rNF3AgAAAIBkPQz7OnbsmEGDBmXw4MG5+OKL07179/zf//1fHnvssXTr1i0DBgzI6aefnl133TXDhg3LMccck8aNG+fVV1/NI488kquuuiqdOnVK//79c9xxx+WXv/xl6tWrl+HDh6dRo0bFccaNG5clS5Zkl112SXl5eW6++eY0atQoW2yxRS2+PQAAAAClbL37Zl+S3HDDDRk8eHB+/OMfp1OnTtl///3z4osvpm3btkmSbt265cknn8xrr72W3XffPd27d89ZZ51V3NDj4z423XTT9OnTJwceeGCOPfbYbLzxxsX7zZs3z7XXXpvevXunW7duefTRR/OHP/whLVq0+NLfFwAAAID1Q6G6urq6totgWVVVVWnWrFnaDL89dcrKa7scAABgPTNr7IDaLgGAT/g4K6qsrExFRcUK262XM/sAAAAAoBQJ+wAAAACgRKx3G3R81Uwd1W+lUzMBAAAA4GNm9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQImoV9sFsHLbnf1Q6pSV13YZAADASswaO6C2SwCAJGb2AQAAAEDJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPZ9hlmzZqVQKGTSpEkrbDNhwoQUCoW8//77X1pdAAAAAPBpwr41YLfddsvcuXPTrFmz2i4FAAAAgPXYeh32LVq0aI3006BBg7Rq1SqFQmGN9AcAAAAAn8dXLuxbunRpxowZky233DKNGjXK9ttvnzvuuCNJ8t5772XQoEFp2bJlGjVqlA4dOuSGG25I8v8tx73tttvSp0+fNGzYMLfcckuWLl2a0aNHZ/PNN09ZWVl22GGHPPjgg8uM+/e//z277bZbGjZsmO222y5PPvlk8d6nl/GOGzcuzZs3z0MPPZTOnTunSZMm6d+/f+bOnbv2fyAAAAAA1lv1aruA1TVmzJjcfPPN+dWvfpUOHTrkqaeeyne/+920bNkyv//97/Pqq6/mT3/6UzbaaKO8/vrr+c9//lPj+Z/85Ce5+OKL07179zRs2DCXX355Lr744lxzzTXp3r17rr/++nz729/O3/72t3To0KH43KmnnprLLrssXbp0ySWXXJKBAwfmzTffTIsWLZZb54IFC3LRRRflpptuSp06dfLd7343I0aMyC233LLc9gsXLszChQuL51VVVWvg1wIAAABgffKVCvsWLlyY888/P48++mh69eqVJGnfvn2eeeaZXHPNNZk/f366d++enj17JknatWu3TB/Dhw/PgQceWDy/6KKLcvrpp+fwww9PklxwwQV54oknctlll+UXv/hFsd2wYcNy0EEHJUl++ctf5sEHH8xvfvObnHbaacutddGiRfnVr36Vrbbaqvj86NGjV/huY8aMyahRo1bj1wAAAACAmr5Sy3hff/31LFiwIN/85jfTpEmT4nHjjTdm5syZ+eEPf5hbb701O+ywQ0477bQ8++yzy/TxcRCYfDR77p///Gd69+5do03v3r0zbdq0Gtc+DheTpF69eunZs+cybT6pvLy8GPQlSevWrTNv3rwVth85cmQqKyuLx5w5c1b8QwAAAADAcnylZvbNnz8/SXL//fdns802q3GvrKwsbdq0yVtvvZUHHnggjzzySPbcc8+ccMIJueiii4rtGjdu/KXUWr9+/RrnhUIh1dXVK2xfVlaWsrKytV0WAAAAACXsKzWzr0uXLikrK8vs2bOz9dZb1zjatGmTJGnZsmWOOuqo3Hzzzbnsssvy61//eoX9VVRUZNNNN83EiRNrXJ84cWK6dOlS49qf//zn4t+LFy/Oyy+/nM6dO6/BtwMAAACAL+YrNbOvadOmGTFiRE4++eQsXbo0X/va11JZWZmJEyemoqIiM2fOzI477phtt902CxcuzB//+MfPDOROPfXUnH322dlqq62yww475IYbbsikSZOW2UjjF7/4RTp06JDOnTvn0ksvzXvvvZfvf//7a/N1AQAAAGC1fKXCviQ599xz07Jly4wZMyZvvPFGmjdvnh49euR//ud/MmfOnIwcOTKzZs1Ko0aNsvvuu+fWW29daX8nnnhiKisr8+Mf/zjz5s1Lly5dct9999XYiTdJxo4dm7Fjx2bSpEnZeuutc99992WjjTZam68KAAAAAKulUL2yD8lRa6qqqtKsWbO0GX576pSV13Y5AADASswaO6C2SwCgxH2cFVVWVqaiomKF7b5S3+wDAAAAAFZM2AcAAAAAJULYBwAAAAAl4iu3Qcf6Zuqofitdhw0AAAAAHzOzDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASkS92i6Aldvu7IdSp6y8tssAAACSzBo7oLZLAICVMrMPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAErEGgv7qqurc+yxx2bDDTdMoVBI8+bNM3z48OL9du3a5bLLLvvC4/z6179OmzZtUqdOnTXS3/KMGzcuzZs3X6N9DhkyJPvvv/8a7RMAAAAAPmmN7cb74IMPZty4cZkwYULat2+fOnXqpFGjRmuq+yRJVVVVhg0blksuuSQHHXRQmjVr9oX7bNeuXYYPH14jmDzssMOyzz77fOG+AQAAAODLtMbCvpkzZ6Z169bZbbfd1lSXy5g9e3YWLVqUAQMGpHXr1mttnEaNGq3xoBIAAAAA1rY1sox3yJAh+dGPfpTZs2enUCikXbt26du3b43Zcp9WKBRyzTXXZN999015eXk6d+6c5557Lq+//nr69u2bxo0bZ7fddsvMmTOTfLS0tmvXrkmS9u3bp1AoZNasWUmSe++9Nz169EjDhg3Tvn37jBo1KosXL07y0fLic845J23btk1ZWVk23XTTnHjiiUmSvn375q233srJJ5+cQqGQQqFQHOvjZbyvvfZaCoVC/v73v9eo/9JLL81WW22VJFmyZEmOPvrobLnllmnUqFE6deqUyy+/fE38tAAAAACwytZI2Hf55Zdn9OjR2XzzzTN37ty8+OKLq/Tcueeem8GDB2fSpEnZZptt8p3vfCfHHXdcRo4cmZdeeinV1dUZNmxYko+W1j766KNJkhdeeCFz585NmzZt8vTTT2fw4ME56aST8uqrr+aaa67JuHHj8rOf/SxJcuedd+bSSy/NNddckxkzZuSee+4phoZ33XVXNt9884wePTpz587N3Llzl6mxY8eO6dmzZ2655ZYa12+55ZZ85zvfSZIsXbo0m2++eX7/+9/n1VdfzVlnnZX/+Z//ye23377Kv+HChQtTVVVV4wAAAACA1bFGwr5mzZqladOmqVu3blq1apWWLVuu0nPf+973cuihh6Zjx445/fTTM2vWrAwaNCj9+vVL586dc9JJJ2XChAlJPlpa26JFiyRJy5Yt06pVq9StWzejRo3KT37ykxx11FFp3759vvnNb+bcc8/NNddck+Sjpb+tWrXKXnvtlbZt22bnnXfO0KFDkyQbbrhh6tatm6ZNm6ZVq1Zp1arVcuscNGhQfve73xXPX3vttbz88ssZNGhQkqR+/foZNWpUevbsmS233DKDBg3K9773vdUK+8aMGZNmzZoVjzZt2qzyswAAAACQrMHdeD+Pbt26Ff/eZJNNkqQ46+7ja//9739XOstt8uTJGT16dJo0aVI8hg4dmrlz52bBggU55JBD8p///Cft27fP0KFDc/fddxeX+K6qww8/PLNmzcqf//znJB/N6uvRo0e22WabYptf/OIX2XHHHdOyZcs0adIkv/71rzN79uxVHmPkyJGprKwsHnPmzFmtGgEAAACgVsO++vXrF//++Ht5y7u2dOnSFfYxf/78jBo1KpMmTSoeU6ZMyYwZM9KwYcO0adMm06dPz9VXX51GjRrl+OOPz9e//vUsWrRolets1apV9thjj/z2t79Nkvz2t78tzupLkltvvTUjRozI0UcfnYcffjiTJk3K9773vXz44YerPEZZWVkqKipqHAAAAACwOtbYbry1pUePHpk+fXq23nrrFbZp1KhRBg4cmIEDB+aEE07INttskylTpqRHjx5p0KBBlixZ8pnjDBo0KKeddlqOOOKIvPHGGzn88MOL9yZOnJjddtstxx9/fPHaxxuLAAAAAMCX5Ssf9p111lnZd99907Zt2xx88MGpU6dOJk+enKlTp+a8887LuHHjsmTJkuyyyy4pLy/PzTffnEaNGmWLLbZIkrRr1y5PPfVUDj/88JSVlWWjjTZa7jgHHnhgfvjDH+aHP/xhvvGNb2TTTTct3uvQoUNuvPHGPPTQQ9lyyy1z00035cUXX8yWW275pfwGAAAAAJDU8jLeNaFfv3754x//mIcffjg77bRTdt1111x66aXFMK958+a59tpr07t373Tr1i2PPvpo/vCHPxQ3+xg9enRmzZqVrbbaaqUbizRt2jQDBw7M5MmTayzhTZLjjjsuBx54YA477LDssssueeedd2rM8gMAAACAL0Ohurq6uraLYFlVVVUf7co7/PbUKSuv7XIAAIAks8YOqO0SAFhPfZwVVVZWrnSvh6/8zD4AAAAA4CPCPgAAAAAoEV/5DTpK3dRR/VY6NRMAAAAAPmZmHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlIh6tV0AK7fd2Q+lTll5bZcBAMB6YtbYAbVdAgDwBZjZBwAAAAAlQtgHAAAAACVC2AcAAAAAJaLkwr7bbrst99xzT22XAQAAAABfupIK+yZMmJAzzjgju+66a22XAgAAAABfupIJ+95+++0MGzYsf/jDH9KqVasvbdx27drlsssu+9LGAwAAAIAVqVfbBawpG220UaZOnVrbZQAAAABArVnnZvb17ds3P/rRjzJ8+PBssMEG2WSTTXLttdfmgw8+yPe+9700bdo0W2+9df70pz8lSZYsWZKjjz46W265ZRo1apROnTrl8ssvr9HnhAkTsvPOO6dx48Zp3rx5evfunbfeeqt4/957702PHj3SsGHDtG/fPqNGjcrixYuTJNXV1TnnnHPStm3blJWVZdNNN82JJ55YrPWtt97KySefnEKhkEKhkCR55513csQRR2SzzTZLeXl5unbtmt/97ndfxs8HAAAAwHpsnQv7kmT8+PHZaKON8sILL+RHP/pRfvjDH+aQQw7Jbrvtlr/85S/51re+lSOPPDILFizI0qVLs/nmm+f3v/99pk2bllGjRuWMM87I7bffniRZvHhx9t9///Tp0yevvPJKnnvuuRx77LHFYO7pp5/O4MGDc9JJJ+XVV1/NNddck3HjxuVnP/tZkuTOO+/MpZdemmuuuSYzZszIPffck65duyZJ7rrrrmy++eYZPXp05s6dm7lz5yZJ/vvf/2bHHXfM/fffn6lTp+bYY4/NkUcemRdeeGGF77xw4cJUVVXVOAAAAABgdRSqq6ura7uIT+rbt2+WLFmSp59+OslHM/eaNWuWAw88MDfeeGOS5H//93/TunXrPPfcc8vdjONHP/pR5s6dmzvuuCPvvvtuWrRokQkTJqRPnz7LtN1rr72y5557ZuTIkcVrN998c0477bT885//zCWXXJJrrrkmU6dOTf369Zd5vl27dhk+fHiGDx++0vfad999s8022+Siiy5a7v1zzjkno0aNWuZ6m+G3p05Z+Ur7BgCANWXW2AG1XQIAsBxVVVVp1qxZKisrU1FRscJ26+TMvm7duhX/rlu3blq0aFGcTZckm2yySZJk3rx5SZKLLroo22yzTRo1apRCoZCrrroqs2fPTpJsuOGGGTJkSPr165eBAwfm8ssvL87AS5LJkydn9OjRadKkSfEYOnRo5s6dmwULFuSQQw7Jf/7zn7Rv3z5Dhw7N3XffXVziuyJLlizJueeem65du2bDDTdMkyZN8tBDDxVrWp6RI0emsrKyeMyZM2f1fzgAAAAA1mvrZNj36Rl0hUKhxrWPl+AuXbo0t9xyS84999xceuml+de//pXq6uqcdtpp+fDDD4vtb7jhhjz33HPZbbfdctttt6Vjx47585//nCSZP39+Ro0alUmTJhWPKVOmZMaMGWnYsGHatGmT6dOn5+qrr06jRo1y/PHH5+tf/3oWLVq0wvovvPDCXH755Tn99NPzxBNPZNKkSenXr1+Nmj6trKwsFRUVNQ4AAAAAWB1f+d14n3vuuey8887Ze++9i9eeffbZZdp179493bt3z8iRI9OrV6/89re/za677poePXpk+vTp2XrrrVc4RqNGjTJw4MAMHDgwJ5xwQrbZZptMmTIlPXr0SIMGDbJkyZIa7SdOnJj99tsv3/3ud5N8FEq+9tpr6dKlyxp6awAAAABY1lc+7OvUqVPGjx+fP/3pT9lqq61y/fXXZ8qUKWnXrl2S5M0338yvf/3rfPvb386mm26a6dOnZ8aMGRk8eHCS5Kyzzsq+++6btm3b5uCDD06dOnUyefLkTJ06Needd17GjRuXJUuWZJdddkl5eXluvvnmNGrUKFtssUWSj77Z99RTT+Xwww9PWVlZNtpoo3To0CF33HFHnn322WywwQa55JJL8q9//UvYBwAAAMBatU4u410dxx13XA499NB85zvfyS677JKqqqocf/zxxfvl5eX5+9//noMOOigdO3bMsccemxNOOCHHHXdckqRfv3754x//mIcffjg77bRTdt1111x66aXFMK958+a59tpr07t373Tr1i2PPvpo/vCHP6RFixZJktGjR2fWrFnZaqut0rJlyyTJT3/60/To0SP9+vVL375906pVq+y///5f7g8DAAAAwHpnnduNl498vMOK3XgBAPgy2Y0XANZNX+ndeAEAAACA1SfsAwAAAIASIewDAAAAgBLxld+Nt9RNHdVvpeuwAQAAAOBjZvYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJEPYBAAAAQIkQ9gEAAABAiRD2AQAAAECJqFfbBbBy2539UOqUldd2GQAAJWXW2AG1XQIAwFphZh8AAAAAlAhhHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhh3wr07ds3w4cPX+H9du3a5bLLLlvl/s4555zssMMOX7guAAAAAFgRYR8AAAAAlAhhHwAAAACUCGHfSixevDjDhg1Ls2bNstFGG+XMM89MdXX1ctvOnj07++23X5o0aZKKiooceuih+de//vUlVwwAAADA+kzYtxLjx49PvXr18sILL+Tyyy/PJZdckuuuu26ZdkuXLs1+++2Xd999N08++WQeeeSRvPHGGznssMNWeayFCxemqqqqxgEAAAAAq6NebRewLmvTpk0uvfTSFAqFdOrUKVOmTMmll16aoUOH1mj32GOPZcqUKXnzzTfTpk2bJMmNN96YbbfdNi+++GJ22mmnzxxrzJgxGTVq1Fp5DwAAAADWD2b2rcSuu+6aQqFQPO/Vq1dmzJiRJUuW1Gg3bdq0tGnTphj0JUmXLl3SvHnzTJs2bZXGGjlyZCorK4vHnDlz1sxLAAAAALDeMLNvHVFWVpaysrLaLgMAAACArzAz+1bi+eefr3H+5z//OR06dEjdunVrXO/cuXPmzJlTYzbeq6++mvfffz9dunT5UmoFAAAAAGHfSsyePTunnHJKpk+fnt/97ne58sorc9JJJy3Tbq+99krXrl0zaNCg/OUvf8kLL7yQwYMHp0+fPunZs2ctVA4AAADA+sgy3pUYPHhw/vOf/2TnnXdO3bp1c9JJJ+XYY49dpl2hUMi9996bH/3oR/n617+eOnXqpH///rnyyitroWoAAAAA1leF6urq6tougmVVVVWlWbNmaTP89tQpK6/tcgAASsqssQNquwQAgNXycVZUWVmZioqKFbazjBcAAAAASoSwDwAAAABKhG/2reOmjuq30qmZAAAAAPAxM/sAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBE1KvtAli57c5+KHXKymu7DACAzBo7oLZLAADgM5jZBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAl4ksN+/r27Zvhw4ev1THatWuXyy67bJXbT5gwIYVCIe+///5aqwkAAAAAvgwltxvviy++mMaNG69y+9122y1z585Ns2bN1mJVAAAAALD2lVzY17Jly9Vq36BBg7Rq1WotVQMAAAAAX561toz3gw8+yODBg9OkSZO0bt06F198cY377733XgYPHpwNNtgg5eXl2XvvvTNjxozi/XHjxqV58+b54x//mE6dOqW8vDwHH3xwFixYkPHjx6ddu3bZYIMNcuKJJ2bJkiXF5z69jLdQKOS6667LAQcckPLy8nTo0CH33Xdf8f7ylvGOGzcubdu2TXl5eQ444IBcfPHFad68efH+kCFDsv/++9d4n+HDh6dv377F86VLl2bMmDHZcsst06hRo2y//fa54447Pt+PCQAAAACrYK2FfaeeemqefPLJ3HvvvXn44YczYcKE/OUvfyneHzJkSF566aXcd999ee6551JdXZ199tknixYtKrZZsGBBrrjiitx666158MEHM2HChBxwwAF54IEH8sADD+Smm27KNddc85kh2qhRo3LooYfmlVdeyT777JNBgwbl3XffXW7b559/PkcffXSGDRuWSZMm5Rvf+EbOO++81X7/MWPG5MYbb8yvfvWr/O1vf8vJJ5+c7373u3nyySeX237hwoWpqqqqcQAAAADA6lgry3jnz5+f3/zmN7n55puz5557JknGjx+fzTffPEkyY8aM3HfffZk4cWJ22223JMktt9ySNm3a5J577skhhxySJFm0aFF++ctfZquttkqSHHzwwbnpppvyr3/9K02aNEmXLl3yjW98I0888UQOO+ywFdYzZMiQHHHEEUmS888/P1dccUVeeOGF9O/ff5m2l19+efr375/TTjstSdKxY8c8++yzefDBB1f5/RcuXJjzzz8/jz76aHr16pUkad++fZ555plcc8016dOnzzLPjBkzJqNGjVrlMQAAAADg09bKzL6ZM2fmww8/zC677FK8tuGGG6ZTp05JkmnTpqVevXo17rdo0SKdOnXKtGnTitfKy8uLQV+SbLLJJmnXrl2aNGlS49q8efNWWk+3bt2Kfzdu3DgVFRUrfGbatGk16kpSDOxW1euvv54FCxbkm9/8Zpo0aVI8brzxxsycOXO5z4wcOTKVlZXFY86cOas1JgAAAACs0xt01K9fv8Z5oVBY7rWlS5eudj+f9czK1KlTJ9XV1TWufXL58fz585Mk999/fzbbbLMa7crKypbbZ1lZ2QrvAQAAAMCqWCsz+7baaqvUr18/zz//fPHae++9l9deey1J0rlz5yxevLjG/XfeeSfTp09Ply5d1kZJq6xz58416kqSP//5zzXOW7Zsmblz59a4NmnSpOLfXbp0SVlZWWbPnp2tt966xtGmTZu1VjsAAAAA67e1MrOvSZMmOfroo3PqqaemRYsW2XjjjXPGGWekTp2PssUOHTpkv/32y9ChQ3PNNdekadOm+clPfpLNNtss++2339ooaZWdeOKJ6d27dy666KLst99+eeihh5b5Xt8ee+yRCy+8MDfeeGN69eqVm2++OVOnTk337t2TJE2bNs2IESNy8sknZ+nSpfna176WysrKTJw4MRUVFTnqqKNq49UAAAAAKHFrbTfeCy+8MLvvvnsGDhyYvfbaK1/72tey4447Fu/fcMMN2XHHHbPvvvumV69eqa6uzgMPPLDMktsv26677pprr702l19+ebbffvs8/PDD+elPf1qjTb9+/XLmmWfmtNNOy0477ZR///vfGTx4cI025557bs4888yMGTMmnTt3Tv/+/XP//fdnyy23/DJfBwAAAID1SKH60x+fYxnjxo3L8OHD8/77739pY1ZVVaVZs2ZpM/z21Ckr/9LGBQBYkVljB9R2CQAA662Ps6LKyspUVFSssN1am9kHAAAAAHy5hH0AAAAAUCKEfatgyJAhX+oSXgAAAAD4PNbKbrysOVNH9VvpOmwAAAAA+JiZfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCLq1XYBrNx2Zz+UOmXltV0GACVg1tgBtV0CAACwlpnZBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtj3OQ0ZMiT7779/bZcBAAAAAEXrfdgntAMAAACgVKz3YR8AAAAAlIr1Juy744470rVr1zRq1CgtWrTIXnvtlVNPPTXjx4/Pvffem0KhkEKhkAkTJiRJpkyZkj322KPY/thjj838+fNX2P+LL76Yli1b5oILLkiSvP/++znmmGPSsmXLVFRUZI899sjkyZO/jFcFAAAAYD1Vr7YL+DLMnTs3RxxxRH7+85/ngAMOyL///e88/fTTGTx4cGbPnp2qqqrccMMNSZINN9wwH3zwQfr165devXrlxRdfzLx583LMMcdk2LBhGTdu3DL9P/744znwwAPz85//PMcee2yS5JBDDkmjRo3ypz/9Kc2aNcs111yTPffcM6+99lo23HDDZfpYuHBhFi5cWDyvqqpaOz8GAAAAACVrvQn7Fi9enAMPPDBbbLFFkqRr165JkkaNGmXhwoVp1apVsf348ePz3//+NzfeeGMaN26cJLnqqqsycODAXHDBBdlkk02Kbe++++4MHjw41113XQ477LAkyTPPPJMXXngh8+bNS1lZWZLkoosuyj333JM77rijGAh+0pgxYzJq1Ki18wMAAAAAsF5YL5bxbr/99tlzzz3TtWvXHHLIIbn22mvz3nvvrbD9tGnTsv322xeDviTp3bt3li5dmunTpxevPf/88znkkENy0003FYO+JJk8eXLmz5+fFi1apEmTJsXjzTffzMyZM5c75siRI1NZWVk85syZswbeHAAAAID1yXoxs69u3bp55JFH8uyzz+bhhx/OlVdemTPOOCPPP//8F+p3q622SosWLXL99ddnwIABqV+/fpJk/vz5ad26dfH7f5/UvHnz5fZVVlZWnAUIAAAAAJ/HejGzL0kKhUJ69+6dUaNG5a9//WsaNGiQu+++Ow0aNMiSJUtqtO3cuXMmT56cDz74oHht4sSJqVOnTjp16lS8ttFGG+Xxxx/P66+/nkMPPTSLFi1KkvTo0SP/+7//m3r16mXrrbeucWy00UZfzgsDAAAAsN5ZL8K+559/Pueff35eeumlzJ49O3fddVf+7//+L507d067du3yyiuvZPr06Xn77bezaNGiDBo0KA0bNsxRRx2VqVOn5oknnsiPfvSjHHnkkTW+15ckG2+8cR5//PH8/e9/zxFHHJHFixdnr732Sq9evbL//vvn4YcfzqxZs/Lss8/mjDPOyEsvvVRLvwIAAAAApW69CPsqKiry1FNPZZ999knHjh3z05/+NBdffHH23nvvDB06NJ06dUrPnj3TsmXLTJw4MeXl5XnooYfy7rvvZqeddsrBBx+cPffcM1ddddVy+2/VqlUef/zxTJkyJYMGDcrSpUvzwAMP5Otf/3q+973vpWPHjjn88MPz1ltvLRMWAgAAAMCaUqiurq6u7SJYVlVVVZo1a5Y2w29PnbLy2i4HgBIwa+yA2i4BAAD4nD7OiiorK1NRUbHCduvFzD4AAAAAWB8I+wAAAACgRNSr7QJYuamj+q10aiYAAAAAfMzMPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBH1arsAVm67sx9KnbLy2i4DgHXIrLEDarsEAABgHWVmHwAAAACUCGEfAAAAAJQIYR8AAAAAlAhhHwAAAACUiK9s2Ne3b98MHz68eL5gwYIcdNBBqaioSKFQyPvvv/+ZfUyYMGGV2wIAAADAuu4rG/Z92vjx4/P000/n2Wefzdy5c9OsWbMvbex27drlsssu+9LGAwAAAIDlqVfbBawpM2fOTOfOnbPddtvVdikAAAAAUCvWysy+Bx98MF/72tfSvHnztGjRIvvuu29mzpyZJJk1a1YKhULuuuuufOMb30h5eXm23377PPfcc8Xn33nnnRxxxBHZbLPNUl5enq5du+Z3v/vdCsfr27dvLr744jz11FMpFArp27dvkuSmm25Kz54907Rp07Rq1Srf+c53Mm/evGWenzhxYrp165aGDRtm1113zdSpU2vcf+aZZ7L77runUaNGadOmTU488cR88MEHxbHfeuutnHzyySkUCikUCsXn7rzzzmy77bYpKytLu3btcvHFF3/u3xQAAAAAPstaCfs++OCDnHLKKXnppZfy2GOPpU6dOjnggAOydOnSYpszzjgjI0aMyKRJk9KxY8ccccQRWbx4cZLkv//9b3bcccfcf//9mTp1ao499tgceeSReeGFF5Y73l133ZWhQ4emV69emTt3bu66664kyaJFi3Luuedm8uTJueeeezJr1qwMGTJkmedPPfXUXHzxxXnxxRfTsmXLDBw4MIsWLUry0YzB/v3756CDDsorr7yS2267Lc8880yGDRtWHHvzzTfP6NGjM3fu3MydOzdJ8vLLL+fQQw/N4YcfnilTpuScc87JmWeemXHjxi33HRYuXJiqqqoaBwAAAACsjkJ1dXX12h7k7bffTsuWLTNlypQ0adIkW265Za677rocffTRSZJXX3012267baZNm5ZtttlmuX3su+++2WabbXLRRRcl+WhG3Q477FD8Vt7w4cMzadKkTJgwYYV1vPTSS9lpp53y73//O02aNMmECRPyjW98I7feemsOO+ywJMm7776bzTffPOPGjcuhhx6aY445JnXr1s0111xT7OeZZ55Jnz598sEHH6Rhw4Zp165dhg8fXmPDkEGDBuX//u//8vDDDxevnXbaabn//vvzt7/9bZnazjnnnIwaNWqZ622G3546ZeUrfCcA1j+zxg6o7RIAAIAvWVVVVZo1a5bKyspUVFSssN1amdk3Y8aMHHHEEWnfvn0qKirSrl27JMns2bOLbbp161b8u3Xr1klSXGK7ZMmSnHvuuenatWs23HDDNGnSJA899FCN51fFyy+/nIEDB6Zt27Zp2rRp+vTps0wdSdKrV6/i3xtuuGE6deqUadOmJUkmT56ccePGpUmTJsWjX79+Wbp0ad58880Vjj1t2rT07t27xrXevXtnxowZWbJkyTLtR44cmcrKyuIxZ86c1XpXAAAAAFgrG3QMHDgwW2yxRa699tpsuummWbp0abbbbrt8+OGHxTb169cv/v3xd+4+XuZ74YUX5vLLL89ll12Wrl27pnHjxhk+fHiN5z/LBx98kH79+qVfv3655ZZb0rJly8yePTv9+vVbrX7mz5+f4447LieeeOIy99q2bbvK/XyWsrKylJWVrbH+AAAAAFj/rPGw75133sn06dNz7bXXZvfdd0/y0bLX1TFx4sTst99++e53v5vkoxDwtddeS5cuXVa5j7///e955513Mnbs2LRp0ybJR8t4l+fPf/5zMbh777338tprr6Vz585Jkh49euTVV1/N1ltvvcKxGjRosMxsvc6dO2fixInLvFfHjh1Tt27dVX4PAAAAAFhVa3wZ7wYbbJAWLVrk17/+dV5//fU8/vjjOeWUU1arjw4dOuSRRx7Js88+m2nTpuW4447Lv/71r9Xqo23btmnQoEGuvPLKvPHGG7nvvvty7rnnLrft6NGj89hjj2Xq1KkZMmRINtpoo+y///5JktNPPz3PPvtshg0blkmTJmXGjBm59957ixt0JEm7du3y1FNP5R//+EfefvvtJMmPf/zjPPbYYzn33HPz2muvZfz48bnqqqsyYsSI1XoPAAAAAFhVazzsq1OnTm699da8/PLL2W677XLyySfnwgsvXK0+fvrTn6ZHjx7p169f+vbtm1atWhXDt1XVsmXLjBs3Lr///e/TpUuXjB07tri5x6eNHTs2J510Unbcccf87//+b/7whz+kQYMGST76tuCTTz6Z1157Lbvvvnu6d++es846K5tuumnx+dGjR2fWrFnZaqut0rJlyyQfzQi8/fbbc+utt2a77bbLWWedldGjRy93N2AAAAAAWBO+lN14WX0f77BiN14APs1uvAAAsP6p1d14AQAAAIAvn7APAAAAAEqEsA8AAAAASkS92i6AlZs6qt9K12EDAAAAwMfM7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEiHsAwAAAIASIewDAAAAgBJRr7YLYOW2O/uh1Ckrr+0yAFgNs8YOqO0SAACA9ZSZfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9q6Fdu3a57LLLarsMAAAAAFiu9TLsKxQKueeee2q7DAAAAABYo9bLsA8AAAAAStE6Gfb9+9//zqBBg9K4ceO0bt06l156afr27Zvhw4cnWf7MvObNm2fcuHFJkg8//DDDhg1L69at07Bhw2yxxRYZM2ZMko+W4ibJAQcckEKhUDyfOXNm9ttvv2yyySZp0qRJdtpppzz66KMrrfOSSy5J165d07hx47Rp0ybHH3985s+fX7z/1ltvZeDAgdlggw3SuHHjbLvttnnggQe+8O8DAAAAAMuzToZ9p5xySiZOnJj77rsvjzzySJ5++un85S9/WeXnr7jiitx33325/fbbM3369Nxyyy3FUO/FF19Mktxwww2ZO3du8Xz+/PnZZ5998thjj+Wvf/1r+vfvn4EDB2b27NkrHKdOnTq54oor8re//S3jx4/P448/ntNOO614/4QTTsjChQvz1FNPZcqUKbngggvSpEmT5fa1cOHCVFVV1TgAAAAAYHXUq+0CPu3f//53xo8fn9/+9rfZc889k3wUzG266aar3Mfs2bPToUOHfO1rX0uhUMgWW2xRvNeyZcskH80EbNWqVfH69ttvn+233754fu655+buu+/Offfdl2HDhi13nI9nGiYfzRg877zz8oMf/CBXX311sY6DDjooXbt2TZK0b99+hTWPGTMmo0aNWuV3BAAAAIBPW+dm9r3xxhtZtGhRdt555+K1Zs2apVOnTqvcx5AhQzJp0qR06tQpJ554Yh5++OHPfGb+/PkZMWJEOnfunObNm6dJkyaZNm3aSmf2Pfroo9lzzz2z2WabpWnTpjnyyCPzzjvvZMGCBUmSE088Meedd1569+6ds88+O6+88soK+xo5cmQqKyuLx5w5c1b5fQEAAAAgWQfDvlVRKBRSXV1d49qiRYuKf/fo0SNvvvlmzj333PznP//JoYcemoMPPnilfY4YMSJ33313zj///Dz99NOZNGlSunbtmg8//HC57WfNmpV999033bp1y5133pmXX345v/jFL5Kk+MwxxxyTN954I0ceeWSmTJmSnj175sorr1xuf2VlZamoqKhxAAAAAMDqWOfCvvbt26d+/frFb+klSWVlZV577bXiecuWLTN37tzi+YwZM4qz6T5WUVGRww47LNdee21uu+223HnnnXn33XeTJPXr18+SJUtqtJ84cWKGDBmSAw44IF27dk2rVq0ya9asFdb58ssvZ+nSpbn44ouz6667pmPHjvnnP/+5TLs2bdrkBz/4Qe666678+Mc/zrXXXrtavwcAAAAArKp17pt9TZs2zVFHHZVTTz01G264YTbeeOOcffbZqVOnTgqFQpJkjz32yFVXXZVevXplyZIlOf3001O/fv1iH5dccklat26d7t27p06dOvn973+fVq1apXnz5kk++r7eY489lt69e6esrCwbbLBBOnTokLvuuisDBw5MoVDImWeemaVLl66wzq233jqLFi3KlVdemYEDB2bixIn51a9+VaPN8OHDs/fee6djx45577338sQTT6Rz585r/kcDAAAAgKyDM/uSj8K6Xr16Zd99981ee+2V3r17p3PnzmnYsGGS5OKLL06bNm2y++675zvf+U5GjBiR8vLy4vNNmzbNz3/+8/Ts2TM77bRTZs2alQceeCB16tQpPv/II4+kTZs26d69e3HMDTbYILvttlsGDhyYfv36pUePHiuscfvtt88ll1ySCy64INttt11uueWWjBkzpkabJUuW5IQTTkjnzp3Tv3//dOzYsbh5BwAAAACsaYXqT3/8bh30wQcfZLPNNsvFF1+co48+urbL+VJUVVWlWbNmaTP89tQpK//sBwBYZ8waO6C2SwAAAErMx1lRZWXlSvd6WOeW8SbJX//61/z973/PzjvvnMrKyowePTpJst9++9VyZQAAAACw7lonw74kueiiizJ9+vQ0aNAgO+64Y55++ulstNFGtV0WAAAAAKyzvhLLeNdHqzo1EwAAAIDSt6pZ0Tq5QQcAAAAAsPqEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAi6tV2Aazcdmc/lDpl5bVdBsB6a9bYAbVdAgAAwCozsw8AAAAASoSwDwAAAABKhLAPAAAAAErEVybsmzVrVs4777zMnz+/tksBAAAAgHXSaoV9ffv2zfDhw9dSKSu2cOHCHHLIIdloo43SpEmTL318AAAAAPgqWCdm9g0ZMiT777//Cu+ffPLJ+da3vpUf/OAHX15RAAAAAPAVU682B1+yZEkKhcJntrv66qu/hGoAAAAA4KtttWf2LV26NKeddlo23HDDtGrVKuecc07x3iWXXJKuXbumcePGadOmTY4//vga39gbN25cmjdvnvvuuy9dunRJWVlZvv/972f8+PG59957UygUUigUMmHChCTJnDlzcuihh6Z58+bZcMMNs99++2XWrFnF/j6eEXj++ednk002SfPmzTN69OgsXrw4p556ajbccMNsvvnmueGGG2q8w5QpU7LHHnukUaNGadGiRY499tgadX7c70UXXZTWrVunRYsWOeGEE7Jo0aJim/feey+DBw/OBhtskPLy8uy9996ZMWNG8f5bb72VgQMHZoMNNkjjxo2z7bbb5oEHHljdnxsAAAAAVtlqh33jx49P48aN8/zzz+fnP/95Ro8enUceeeSjzurUyRVXXJG//e1vGT9+fB5//PGcdtppNZ5fsGBBLrjgglx33XX529/+liuuuCKHHnpo+vfvn7lz52bu3LnZbbfdsmjRovTr1y9NmzbN008/nYkTJ6ZJkybp379/Pvzww2J/jz/+eP75z3/mqaeeyiWXXJKzzz47++67bzbYYIM8//zz+cEPfpDjjjsu/+///b8kyQcffJB+/fplgw02yIsvvpjf//73efTRRzNs2LAadT7xxBOZOXNmnnjiiYwfPz7jxo3LuHHjiveHDBmSl156Kffdd1+ee+65VFdXZ5999ikGgieccEIWLlyYp556KlOmTMkFF1yw0u8NLly4MFVVVTUOAAAAAFgdherq6upVbdy3b98sWbIkTz/9dPHazjvvnD322CNjx45dpv0dd9yRH/zgB3n77beTfDSz73vf+14mTZqU7bffvthuyJAhef/993PPPfcUr918880577zzMm3atOJS3w8//DDNmzfPPffck29961sZMmRIJkyYkDfeeCN16nyUW26zzTbZeOON89RTTyX5aKlws2bNct111+Xwww/Ptddem9NPPz1z5sxJ48aNkyQPPPBABg4cmH/+85/ZZJNNiv3OnDkzdevWTZIceuihqVOnTm699dbMmDEjHTt2zMSJE7PbbrslSd555520adMm48ePzyGHHJJu3brloIMOytlnn71Kv+0555yTUaNGLXO9zfDbU6esfJX6AGDNmzV2QG2XAAAAkKqqqjRr1iyVlZWpqKhYYbvVntnXrVu3GuetW7fOvHnzkiSPPvpo9txzz2y22WZp2rRpjjzyyLzzzjtZsGBBsX2DBg2W6WN5Jk+enNdffz1NmzZNkyZN0qRJk2y44Yb573//m5kzZxbbbbvttsWgL0k22WSTdO3atXhet27dtGjRoljjtGnTsv322xeDviTp3bt3li5dmunTp9fo9+Og79PvOW3atNSrVy+77LJL8X6LFi3SqVOnTJs2LUly4okn5rzzzkvv3r1z9tln55VXXlnp+44cOTKVlZXFY86cOZ/5GwEAAADAJ6122Fe/fv0a54VCIUuXLs2sWbOy7777plu3brnzzjvz8ssv5xe/+EWS1Fh226hRo1XalGP+/PnZcccdM2nSpBrHa6+9lu985zsrrWdFNa6J91xVxxxzTN54440ceeSRmTJlSnr27Jkrr7xyhe3LyspSUVFR4wAAAACA1bHaYd+KvPzyy1m6dGkuvvji7LrrrunYsWP++c9/rtKzDRo0yJIlS2pc69GjR2bMmJGNN944W2+9dY2jWbNmn7vOzp07Z/Lkyfnggw+K1yZOnJg6deqkU6dOq9zH4sWL8/zzzxevvfPOO5k+fXq6dOlSvNamTZv84Ac/yF133ZUf//jHufbaaz933QAAAADwWdZY2Lf11ltn0aJFufLKK/PGG2/kpptuyq9+9atVerZdu3Z55ZVXMn369Lz99ttZtGhRBg0alI022ij77bdfnn766bz55puZMGFCTjzxxOJmG5/HoEGD0rBhwxx11FGZOnVqnnjiifzoRz/KkUcemU022WSV+ujQoUP222+/DB06NM8880wmT56c7373u9lss82y3377JUmGDx+ehx56KG+++Wb+8pe/5Iknnkjnzp0/d90AAAAA8FnWWNi3/fbb55JLLskFF1yQ7bbbLrfcckvGjBmzSs8OHTo0nTp1Ss+ePdOyZctMnDgx5eXleeqpp9K2bdsceOCB6dy5c44++uj897///UJLXMvLy/PQQw/l3XffzU477ZSDDz44e+65Z6666qrV6ueGG27IjjvumH333Te9evVKdXV1HnjggeLy3yVLluSEE05I586d079//3Ts2DFXX331564bAAAAAD7Lau3Gy5fn4x1W7MYLULvsxgsAAKwL1tpuvAAAAADAuknYBwAAAAAlQtgHAAAAACWiXm0XwMpNHdXvC21IAgAAAMD6w8w+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEcI+AAAAACgRwj4AAAAAKBHCPgAAAAAoEfVquwBWbruzH0qdsvLaLgNgvTBr7IDaLgEAAOALMbMPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEpEyYZ9EyZMSKFQyPvvv1/bpQAAAADAl6Jkwr6+fftm+PDhtV0GAAAAANSakgn71hUffvhhbZcAAAAAwHqqJMK+IUOG5Mknn8zll1+eQqGQQqGQWbNmJUlefvnl9OzZM+Xl5dltt90yffr0Gs/ee++96dGjRxo2bJj27dtn1KhRWbx4cfH+7Nmzs99++6VJkyapqKjIoYcemn/961/F++ecc0522GGHXHfdddlyyy3TsGHD3HjjjWnRokUWLlxYY6z9998/Rx555Nr7IQAAAABYr5VE2Hf55ZenV69eGTp0aObOnZu5c+emTZs2SZIzzjgjF198cV566aXUq1cv3//+94vPPf300xk8eHBOOumkvPrqq7nmmmsybty4/OxnP0uSLF26NPvtt1/efffdPPnkk3nkkUfyxhtv5LDDDqsx/uuvv54777wzd911VyZNmpRDDjkkS5YsyX333VdsM2/evNx///01xv+khQsXpqqqqsYBAAAAAKujJMK+Zs2apUGDBikvL0+rVq3SqlWr1K1bN0nys5/9LH369EmXLl3yk5/8JM8++2z++9//JklGjRqVn/zkJznqqKPSvn37fPOb38y5556ba665Jkny2GOPZcqUKfntb3+bHXfcMbvssktuvPHGPPnkk3nxxReL43/44Ye58cYb071793Tr1i2NGjXKd77zndxwww3FNjfffHPatm2bvn37LvcdxowZk2bNmhWPj8NKAAAAAFhVJRH2rUy3bt2Kf7du3TrJR7PskmTy5MkZPXp0mjRpUjw+nh24YMGCTJs2LW3atKkRvHXp0iXNmzfPtGnTite22GKLtGzZssa4Q4cOzcMPP5x//OMfSZJx48ZlyJAhKRQKy61z5MiRqaysLB5z5sxZMz8AAAAAAOuNerVdwNpWv3794t8fB21Lly5NksyfPz+jRo3KgQceuMxzDRs2XOUxGjduvMy17t27Z/vtt8+NN96Yb33rW/nb3/6W+++/f4V9lJWVpaysbJXHBAAAAIBPK5mwr0GDBlmyZMlqPdOjR49Mnz49W2+99XLvd+7cOXPmzMmcOXOKs/teffXVvP/+++nSpctn9n/MMcfksssuyz/+8Y/stddeluYCAAAAsFaVTNjXrl27PP/885k1a1aaNGlSnL23MmeddVb23XfftG3bNgcffHDq1KmTyZMnZ+rUqTnvvPOy1157pWvXrhk0aFAuu+yyLF68OMcff3z69OmTnj17fmb/3/nOdzJixIhce+21ufHGG9fEawIAAADACpXMN/tGjBiRunXrpkuXLmnZsmVmz579mc/069cvf/zjH/Pwww9np512yq677ppLL700W2yxRZKPlv3ee++92WCDDfL1r389e+21V9q3b5/bbrttlWpq1qxZDjrooDRp0iT777//F3k9AAAAAPhMherq6uraLqKU7bnnntl2221zxRVXrNZzVVVVH+3KO/z21CkrX0vVAfBJs8YOqO0SAAAAluvjrKiysjIVFRUrbFcyy3jXNe+9914mTJiQCRMm5Oqrr67tcgAAAABYDwj71pLu3bvnvffeywUXXJBOnTrVdjkAAAAArAcs411HrerUTAAAAABK36pmRSWzQQcAAAAArO+EfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAihH0AAAAAUCKEfQAAAABQIoR9AAAAAFAi6tV2Aazcdmc/lDpl5bVdBsAXNmvsgNouAQAAoOSZ2QcAAAAAJULYBwAAAAAlQtgHAAAAACVC2AcAAAAAJULYBwAAAAAlQtgHAAAAACVC2PcJH374YW2XAAAAAACf23od9vXt2zfDhg3L8OHDs9FGG6Vfv34pFAqZNGlSsc3777+fQqGQCRMmFK/dd9996dChQxo2bJhvfOMbGT9+fAqFQt5///188MEHqaioyB133FFjrHvuuSeNGzfOv//97y/p7QAAAABY36zXYV+SjB8/Pg0aNMjEiRPzq1/96jPbv/nmmzn44IOz//77Z/LkyTnuuONyxhlnFO83btw4hx9+eG644YYaz91www05+OCD07Rp0+X2u3DhwlRVVdU4AAAAAGB1rPdhX4cOHfLzn/88nTp1SllZ2We2v+aaa9KpU6dceOGF6dSpUw4//PAMGTKkRptjjjkmDz30UObOnZskmTdvXh544IF8//vfX2G/Y8aMSbNmzYpHmzZtvtB7AQAAALD+We/Dvh133HG12k+fPj077bRTjWs777zzMufbbrttxo8fnyS5+eabs8UWW+TrX//6CvsdOXJkKisri8ecOXNWqy4AAAAAWO/DvsaNGxf/rlPno5+jurq6eG3RokWfq99jjjkm48aNS/LREt7vfe97KRQKK2xfVlaWioqKGgcAAAAArI71Puz7pJYtWyZJcfltkhqbdSRJp06d8tJLL9W49uKLLy7T13e/+9289dZbueKKK/Lqq6/mqKOOWvMFAwAAAMAnCPs+oVGjRtl1110zduzYTJs2LU8++WR++tOf1mhz3HHH5e9//3tOP/30vPbaa7n99tuLM/g+OXNvgw02yIEHHphTTz013/rWt7L55pt/ma8CAAAAwHpI2Pcp119/fRYvXpwdd9wxw4cPz3nnnVfj/pZbbpk77rgjd911V7p165Zf/vKXxd14P73Bx9FHH50PP/xwpRtzAAAAAMCaUq+2C6hNEyZMWOZa586d8+yzz9a49slv+CXJt7/97Xz7298unv/sZz/L5ptvnoYNG9Zo949//CMtWrTIfvvtt+aKBgAAAIAVWK/Dvs/r6quvzk477ZQWLVpk4sSJufDCCzNs2LDi/QULFmTu3LkZO3ZsjjvuuDRo0KAWqwUAAABgfWEZ7+cwY8aM7LfffunSpUvOPffc/PjHP84555xTvP/zn/8822yzTVq1apWRI0fWXqEAAAAArFcK1Z9eo8o6oaqqKs2aNUub4benTll5bZcD8IXNGjugtksAAAD4yvo4K6qsrExFRcUK21nGu46bOqrfSv+BAAAAAPAxy3gBAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBECPsAAAAAoEQI+wAAAACgRAj7AAAAAKBE1KvtAli57c5+KHXKymu7DIBlzBo7oLZLAAAA4FPM7AMAAACAEiHsAwAAAIASIewDAAAAgBIh7AMAAACAEvGVDPv69u2b4cOHf+7nZ82alUKhkEmTJiVJJkyYkEKhkPfff3+N1Lem+wMAAACAVfGVDPsAAAAAgGUJ+wAAAACgRHxlw77Fixdn2LBhadasWTbaaKOceeaZqa6uTpIUCoXcc889Ndo3b94848aNW+X+77zzzmy77bYpKytLu3btcvHFF9e4v3Dhwpx++ulp06ZNysrKsvXWW+c3v/nNcvtasGBB9t577/Tu3dvSXgAAAADWmnq1XcDnNX78+Bx99NF54YUX8tJLL+XYY49N27ZtM3To0C/c98svv5xDDz0055xzTg477LA8++yzOf7449OiRYsMGTIkSTJ48OA899xzueKKK7L99tvnzTffzNtvv71MX++//34GDBiQJk2a5JFHHkl5eflyx1y4cGEWLlxYPK+qqvrC7wEAAADA+uUrG/a1adMml156aQqFQjp16pQpU6bk0ksvXSNh3yWXXJI999wzZ555ZpKkY8eOefXVV3PhhRdmyJAhee2113L77bfnkUceyV577ZUkad++/TL9/O///m8OO+ywdOjQIb/97W/ToEGDFY45ZsyYjBo16gvXDgAAAMD66yu7jHfXXXdNoVAonvfq1SszZszIkiVLvnDf06ZNS+/evWtc6927d7H/SZMmpW7duunTp89K+/nmN7+ZrbfeOrfddttKg74kGTlyZCorK4vHnDlzvvB7AAAAALB++cqGfStTKBSK3+/72KJFi9ZY/40aNVqldgMGDMhTTz2VV1999TPblpWVpaKiosYBAAAAAKvjKxv2Pf/88zXO//znP6dDhw6pW7duWrZsmblz5xbvzZgxIwsWLFjlvjt37pyJEyfWuDZx4sR07NgxdevWTdeuXbN06dI8+eSTK+1n7NixOeqoo7LnnnuuUuAHAAAAAF/EV/abfbNnz84pp5yS4447Ln/5y19y5ZVXFnfM3WOPPXLVVVelV69eWbJkSU4//fTUr19/lfv+8Y9/nJ122innnntuDjvssDz33HO56qqrcvXVVydJ2rVrl6OOOirf//73ixt0vPXWW5k3b14OPfTQGn1ddNFFWbJkSfbYY49MmDAh22yzzZr7EQAAAADgE76yYd/gwYPzn//8JzvvvHPq1q2bk046Kccee2yS5OKLL873vve97L777tl0001z+eWX5+WXX17lvnv06JHbb789Z511Vs4999y0bt06o0ePLu7EmyS//OUv8z//8z85/vjj884776Rt27b5n//5n+X2d+mll9YI/Dp27PiF3h0AAAAAlqdQ/emP27FOqKqqSrNmzdJm+O2pU1Ze2+UALGPW2AG1XQIAAMB64+OsqLKycqV7PXxlv9kHAAAAANQk7AMAAACAEvGV/Wbf+mLqqH4rnZoJAAAAAB8zsw8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEqEsA8AAAAASoSwDwAAAABKhLAPAAAAAEpEvdougJXb7uyHUqesvLbLAErIrLEDarsEAAAA1hIz+wAAAACgRAj7AAAAAKBECPsAAAAAoESsd2HfggULct555+XNN9+s7VIAAAAAYI1a7zboGDp0aJo1a5Ytt9yytksBAAAAgDVqvQr7Lr/88ixYsCA33XRTbZcCAAAAAGvcerWM96STTsrdd9+dOnXW3mtXV1dn8eLFa61/AAAAAFiRdSbse/DBB/O1r30tzZs3T4sWLbLvvvtm5syZxfsvvPBCunfvnoYNG6Znz565++67UygUMmnSpCTJuHHj0rx58xp93nPPPSkUCsXzc845JzvssEPxfMiQIdl///0zatSotGzZMhUVFfnBD36QDz/8sNhm6dKlGTNmTLbccss0atQo22+/fe64447i/QkTJqRQKORPf/pTdtxxx5SVleWZZ57JwoULc+KJJ2bjjTdOw4YN87WvfS0vvvjimv3RAAAAAOAT1pmw74MPPsgpp5ySl156KY899ljq1KmTAw44IEuXLs38+fOz7777pkuXLnn55ZdzzjnnZMSIEWtk3MceeyzTpk3LhAkT8rvf/S533XVXRo0aVbw/ZsyY3HjjjfnVr36Vv/3tbzn55JPz3e9+N08++WSNfn7yk59k7NixmTZtWrp165bTTjstd955Z8aPH5+//OUv2XrrrdOvX7+8++67y61j4cKFqaqqqnEAAAAAwOpYZ77Zd9BBB9U4v/7669OyZcu8+uqrefbZZ7N06dL85je/ScOGDbPtttvm//2//5cf/vCHX3jcBg0a5Prrr095eXm23XbbjB49OqeeemrOPffcLFq0KOeff34effTR9OrVK0nSvn37PPPMM7nmmmvSp0+fYj+jR4/ON7/5zSQfBZe//OUvM27cuOy9995JkmuvvTaPPPJIfvOb3+TUU09dpo4xY8bUCBkBAAAAYHWtM2HfjBkzctZZZ/3/2rv34BrvPI7jn3NcTi7kJIIIRUJYIS6paAa9UDoYW2unu9rUuHQj2FnVom6rkaheskV1y9ZMmdLZ2obuTpldaXWH7nQbYdHEnSWaRt0vJYm2uchv/9jJGaciJRfPyZP3a+aMnOf5Pc/5fY/vPJn55Dnnp127dunSpUsqLy+XJOXn53vulvPz8/OMrwjfaqp3794KCAjwOm9RUZFOnTqloqIifffdd54Qr0JJSYliY2O9tsXFxXl+zs3NVWlpqQYOHOjZ1qRJEz3wwAM6cuRIpfOYP3++Zs6c6XleUFCg9u3b16g2AAAAAAAANCw+E/Y9/vjj6tixo1avXq22bduqvLxcMTExXt+fVxWn0yljjNe20tLSGs2pqKhIkrRlyxa1a9fOa5/L5fJ6HhgYWKPXcrlct5wTAAAAAAAAuBs+8Z19ly9f1rFjx/Tiiy9qyJAhio6O1rfffuvZHx0drf379+uHH37wbNu5c6fXOVq1aqXCwkJdv37ds61i8Y6q7Nu3T99//73XeZs1a6b27dure/fucrlcys/PV1RUlNejqrvuOnfurKZNmyozM9OzrbS0VLt371b37t1/ck4AAAAAAABAdfhE2BcSEqLQ0FC98847OnHihLZv3+71kdann35aDodDSUlJOnz4sDIyMrR06VKvc8THxysgIEC///3vlZubq7/85S9at27dT752SUmJEhMTPedNSUnRtGnT5HQ61bx5c73wwguaMWOG3nvvPeXm5urLL7/UihUr9N577932nIGBgfrtb3+r2bNn65NPPtHhw4eVlJSk7777TomJidV+nwAAAAAAAICq+ETY53Q6lZ6err179yomJkYzZszQkiVLPPubNWumv//97zpw4IBiY2O1YMEC/eEPf/A6R4sWLfT+++8rIyNDPXv21AcffKDU1NSffO0hQ4aoS5cuevjhh/Xkk09q1KhRXsctXrxYycnJeu211xQdHa3hw4dry5YtioyMrPK8aWlpeuKJJzRu3Djdf//9OnHihLZu3aqQkJC7em8AAAAAAACAO+UwP/6iu3oiLy9PkZGRys7OVp8+fap1jokTJ+rq1avatGlTrc6tNhQUFMjtdqv98xvldAX89AEAcIfy0kZaPQUAAAAAwF2qyIquXbumoKCg247ziTv7AAAAAAAAANQcYR8AAAAAAABgE42tnkB1RUREqKafQL6TBTwAAAAAAACA+qLehn0NxcFFw6r8HDYAAAAAAABQgY/xAgAAAAAAADZB2AcAAAAAAADYBGEfAAAAAAAAYBOEfQAAAAAAAIBNEPYBAAAAAAAANkHYBwAAAAAAANgEYR8AAAAAAABgE4R9AAAAAAAAgE0Q9gEAAAAAAAA2QdgHAAAAAAAA2ARhHwAAAAAAAGAThH0AAAAAAACATRD2AQAAAAAAADZB2AcAAAAAAADYBGEfAAAAAAAAYBOEfQAAAAAAAIBNEPYBAAAAAAAANkHYBwAAAAAAANhEY6sngKrFpGyV0xVg9TQA+Ji8tJFWTwEAAAAA4IO4sw8AAAAAAACwCcI+AAAAAAAAwCYI+wAAAAAAAACbIOwDAAAAAAAAbIKwr5ocDoc2bdp0x+NTU1PVp0+fOpsPAAAAAAAAwGq81XT27FmFhIRYPQ0AAAAAAADAg7CvEqWlpWrSpEmVY9q0aXOPZgMAAAAAAADcmWp/jLe8vFyvv/66oqKi5HK51KFDB73yyiuSpAMHDujRRx+Vv7+/QkNDNXnyZBUVFXmOnThxokaPHq1XX31VYWFhCg4O1ksvvaSysjLNnj1bLVq00H333ae1a9d6jsnLy5PD4dDGjRv10EMPyd/fX/369dN///tf7d69W3FxcWrWrJlGjBihixcves11zZo1io6Olp+fn7p166a33377lvNu2LBBjzzyiPz8/LR+/XpJ0rvvvqsePXrI5XIpPDxc06ZN8xz344/xzp07V127dlVAQIA6deqk5ORklZaWVvftBQAAAAAAAO5ate/smz9/vlavXq3ly5frwQcf1NmzZ3X06FFdv35dw4YNU//+/bV7925duHBBkyZN0rRp07Ru3TrP8du3b9d9992nzz//XJmZmUpMTNSOHTv08MMPa9euXdqwYYOmTJmixx57TPfdd5/nuJSUFL355pvq0KGDfvOb3+jpp59W8+bN9cc//lEBAQEaM2aMFi5cqFWrVkmS1q9fr4ULF2rlypWKjY1Vdna2kpKSFBgYqAkTJnjOO2/ePC1btkyxsbHy8/PTqlWrNHPmTKWlpWnEiBG6du2aMjMzb/t+NG/eXOvWrVPbtm114MABJSUlqXnz5pozZ84dvZ/FxcUqLi72PC8oKLjT/woAAAAAAABAkuQwxpi7PaiwsFCtWrXSypUrNWnSJK99q1ev1ty5c3Xq1CkFBgZKkjIyMvT444/rzJkzCgsL08SJE/Wvf/1LJ0+elNP5/5sLu3XrptatW+vzzz+XJN24cUNut1tr1qzRU089pby8PEVGRmrNmjVKTEyUJKWnpyshIUHbtm3To48+KklKS0vTunXrdPToUUlSVFSUFi9erISEBM8cX375ZWVkZGjHjh2e87755pt67rnnPGPatWunZ555Ri+//HLlb5zDoY8++kijR4+udP/SpUuVnp6uPXv2SPr/Ah2bNm1STk5OpeNTU1O1aNGiW7a3f36jnK6ASo8B0HDlpY20egoAAAAAgHuooKBAbrdb165dU1BQ0G3HVevOviNHjqi4uFhDhgypdF/v3r09QZ8kDRw4UOXl5Tp27JjCwsIkST169PAEfZIUFhammJgYz/NGjRopNDRUFy5c8Dp/r169vI6RpJ49e3ptqzjm+vXrys3NVWJiopKSkjxjysrK5Ha7vc4bFxfn+fnChQs6c+ZMpfXdzoYNG/TWW28pNzdXRUVFKisrq/KN/7H58+dr5syZnucFBQVq3779HR8PAAAAAAAAVCvs8/f3r/EL/3gBDIfDUem28vLy2x7ncDgq3VZxTMX3BK5evVrx8fFe52nUqJHX85vDybutLysrS2PHjtWiRYs0bNgwud1upaena9myZXd8DpfLJZfLdVevCwAAAAAAANysWgt0dOnSRf7+/tq2bdst+6Kjo7Vv3z5dv37dsy0zM1NOp1M/+9nPqj/TaggLC1Pbtm118uRJRUVFeT0iIyNve1zz5s0VERFRaX2V2bFjhzp27KgFCxYoLi5OXbp00ddff11bZQAAAAAAAAB3pFp39vn5+Wnu3LmaM2eOmjZtqoEDB+rixYs6dOiQxo4dq5SUFE2YMEGpqam6ePGinn32WY0bN87zsdt7adGiRZo+fbrcbreGDx+u4uJi7dmzR99++63Xx2Z/LDU1VVOnTlXr1q01YsQIFRYWKjMzU88+++wtY7t06aL8/Hylp6erX79+2rJliz766KO6LAsAAAAAAAC4RbVX401OTlbjxo21cOFCnTlzRuHh4Zo6daoCAgK0detWPffcc+rXr58CAgL0xBNP6I033qjNed+xSZMmKSAgQEuWLNHs2bMVGBionj176vnnn6/yuAkTJuiHH37Q8uXL9cILL6hly5b61a9+VenYUaNGacaMGZo2bZqKi4s1cuRIJScnKzU1tfYLAgAAAAAAAG6jWqvxou5VrLDCarwAKsNqvAAAAADQsNzparzV+s4+AAAAAAAAAL6HsA8AAAAAAACwiWp/Zx/ujYOLhlV5ayYAAAAAAABQgTv7AAAAAAAAAJsg7AMAAAAAAABsgrAPAAAAAAAAsAnCPgAAAAAAAMAmCPsAAAAAAAAAmyDsAwAAAAAAAGyCsA8AAAAAAACwCcI+AAAAAAAAwCYI+wAAAAAAAACbIOwDAAAAAAAAbIKwDwAAAAAAALAJwj4AAAAAAADAJgj7AAAAAAAAAJsg7AMAAAAAAABsgrAPAAAAAAAAsAnCPgAAAAAAAMAmCPsAAAAAAAAAmyDsAwAAAAAAAGyisdUTQNViUrbK6QqwehoA7qG8tJFWTwEAAAAAUE9xZx8AAAAAAABgE4R9AAAAAAAAgE0Q9gEAAAAAAAA2QdgHAAAAAAAA2ARhXxXy8vLkcDiUk5Nj9VQAAAAAAACAn0TYBwAAAAAAANgEYZ8FSkpKrJ4CAAAAAAAAbKjBh32ffPKJHnzwQQUHBys0NFQ///nPlZub6zXm5MmTGjx4sAICAtS7d29lZWV57f/b3/6mHj16yOVyKSIiQsuWLfPaHxERocWLF2v8+PEKCgrS5MmT67wuAAAAAAAANDwNPuy7fv26Zs6cqT179mjbtm1yOp365S9/qfLycs+YBQsW6IUXXlBOTo66du2qhIQElZWVSZL27t2rMWPG6KmnntKBAweUmpqq5ORkrVu3zut1li5dqt69eys7O1vJycm3zKO4uFgFBQVeDwAAAAAAAOBuOIwxxupJ+JJLly6pVatWOnDggJo1a6bIyEitWbNGiYmJkqTDhw+rR48eOnLkiLp166axY8fq4sWL+vTTTz3nmDNnjrZs2aJDhw5J+v+dfbGxsfroo49u+7qpqalatGjRLdvbP79RTldALVcJwJflpY20egoAAAAAAB9TUFAgt9uta9euKSgo6LbjGvydfcePH1dCQoI6deqkoKAgRURESJLy8/M9Y3r16uX5OTw8XJJ04cIFSdKRI0c0cOBAr3MOHDhQx48f140bNzzb4uLiqpzH/Pnzde3aNc/j1KlTNaoLAAAAAAAADU9jqydgtccff1wdO3bU6tWr1bZtW5WXlysmJsZrEY0mTZp4fnY4HJLk9THfOxEYGFjlfpfLJZfLdVfnBAAAAAAAAG7WoMO+y5cv69ixY1q9erUeeughSdIXX3xxV+eIjo5WZmam17bMzEx17dpVjRo1qrW5AgAAAAAAAD+lQYd9ISEhCg0N1TvvvKPw8HDl5+dr3rx5d3WOWbNmqV+/flq8eLGefPJJZWVlaeXKlXr77bfraNYAAAAAAABA5Rr0d/Y5nU6lp6dr7969iomJ0YwZM7RkyZK7Osf999+vjRs3Kj09XTExMVq4cKFeeuklTZw4sW4mDQAAAAAAANwGq/H6qIoVVliNF2h4WI0XAAAAAPBjrMYLAAAAAAAANDCEfQAAAAAAAIBNEPYBAAAAAAAANtGgV+OtDw4uGlbl57ABAAAAAACACtzZBwAAAAAAANgEYR8AAAAAAABgE4R9AAAAAAAAgE0Q9gEAAAAAAAA2QdgHAAAAAAAA2ARhHwAAAAAAAGAThH0AAAAAAACATTS2egKonDFGklRQUGDxTAAAAAAAAGC1ioyoIjO6HcI+H3X58mVJUvv27S2eCQAAAAAAAHxFYWGh3G73bfcT9vmoFi1aSJLy8/Or/A8ErFJQUKD27dvr1KlTCgoKsno6gBf6E76M/oQvoz/h6+hR+DL6E3XNGKPCwkK1bdu2ynGEfT7K6fz/1ym63W4uEvBpQUFB9Ch8Fv0JX0Z/wpfRn/B19Ch8Gf2JunQnN4SxQAcAAAAAAABgE4R9AAAAAAAAgE0Q9vkol8ullJQUuVwuq6cCVIoehS+jP+HL6E/4MvoTvo4ehS+jP+ErHOan1usFAAAAAAAAUC9wZx8AAAAAAABgE4R9AAAAAAAAgE0Q9gEAAAAAAAA2QdgHAAAAAAAA2ARhHwAAAAAAAGAThH115E9/+pMiIiLk5+en+Ph4/ec//6ly/Icffqhu3brJz89PPXv2VEZGhtd+Y4wWLlyo8PBw+fv7a+jQoTp+/LjXmCtXrmjs2LEKCgpScHCwEhMTVVRUVOu1wR6s6NFXXnlFAwYMUEBAgIKDg2u7JNjIve7PvLw8JSYmKjIyUv7+/urcubNSUlJUUlJSJ/WhfrPi+jlq1Ch16NBBfn5+Cg8P17hx43TmzJlarw32YEWPViguLlafPn3kcDiUk5NTWyXBRqzoz4iICDkcDq9HWlpardeG+s+q6+eWLVsUHx8vf39/hYSEaPTo0bVZFhoig1qXnp5umjZtat59911z6NAhk5SUZIKDg8358+crHZ+ZmWkaNWpkXn/9dXP48GHz4osvmiZNmpgDBw54xqSlpRm32202bdpk9u3bZ0aNGmUiIyPN999/7xkzfPhw07t3b7Nz507z73//20RFRZmEhIQ6rxf1j1U9unDhQvPGG2+YmTNnGrfbXddlop6yoj8//vhjM3HiRLN161aTm5trNm/ebFq3bm1mzZp1T2pG/WHV9fONN94wWVlZJi8vz2RmZpr+/fub/v3713m9qH+s6tEK06dPNyNGjDCSTHZ2dl2ViXrKqv7s2LGjeemll8zZs2c9j6KiojqvF/WLVf3517/+1YSEhJhVq1aZY8eOmUOHDpkNGzbUeb2wN8K+OvDAAw+Y3/3ud57nN27cMG3btjWvvfZapePHjBljRo4c6bUtPj7eTJkyxRhjTHl5uWnTpo1ZsmSJZ//Vq1eNy+UyH3zwgTHGmMOHDxtJZvfu3Z4xH3/8sXE4HOb06dO1VhvswYoevdnatWsJ+3BbVvdnhddff91ERkbWpBTYkK/05+bNm43D4TAlJSU1KQc2ZGWPZmRkmG7duplDhw4R9qFSVvVnx44dzfLly2uxEtiRFf1ZWlpq2rVrZ9asWVPb5aCB42O8taykpER79+7V0KFDPducTqeGDh2qrKysSo/JysryGi9Jw4YN84z/6quvdO7cOa8xbrdb8fHxnjFZWVkKDg5WXFycZ8zQoUPldDq1a9euWqsP9Z9VPQrcCV/qz2vXrqlFixY1KQc24yv9eeXKFa1fv14DBgxQkyZNaloWbMTKHj1//rySkpL05z//WQEBAbVZFmzC6mtoWlqaQkNDFRsbqyVLlqisrKy2SoMNWNWfX375pU6fPi2n06nY2FiFh4drxIgROnjwYG2XiAaGsK+WXbp0STdu3FBYWJjX9rCwMJ07d67SY86dO1fl+Ip/f2pM69atvfY3btxYLVq0uO3romGyqkeBO+Er/XnixAmtWLFCU6ZMqVYdsCer+3Pu3LkKDAxUaGio8vPztXnz5hrVA/uxqkeNMZo4caKmTp3q9Ydn4GZWXkOnT5+u9PR0ffbZZ5oyZYpeffVVzZkzp8Y1wT6s6s+TJ09KklJTU/Xiiy/qH//4h0JCQjRo0CBduXKl5oWhwSLsAwDgJqdPn9bw4cP161//WklJSVZPB/CYPXu2srOz9emnn6pRo0YaP368jDFWTwvQihUrVFhYqPnz51s9FaBSM2fO1KBBg9SrVy9NnTpVy5Yt04oVK1RcXGz11NDAlZeXS5IWLFigJ554Qn379tXatWvlcDj04YcfWjw71GeEfbWsZcuWatSokc6fP++1/fz582rTpk2lx7Rp06bK8RX//tSYCxcueO0vKyvTlStXbvu6aJis6lHgTljdn2fOnNHgwYM1YMAAvfPOOzWqBfZjdX+2bNlSXbt21WOPPab09HRlZGRo586dNaoJ9mJVj27fvl1ZWVlyuVxq3LixoqKiJElxcXGaMGFCzQuDLVh9Db1ZfHy8ysrKlJeXd7dlwKas6s/w8HBJUvfu3T37XS6XOnXqpPz8/BpUhIaOsK+WNW3aVH379tW2bds828rLy7Vt2zb179+/0mP69+/vNV6S/vnPf3rGR0ZGqk2bNl5jCgoKtGvXLs+Y/v376+rVq9q7d69nzPbt21VeXq74+Phaqw/1n1U9CtwJK/vz9OnTGjRokOcvqk4nvyLhzZeunxV3AnBXCm5mVY++9dZb2rdvn3JycpSTk6OMjAxJ0oYNG/TKK6/Uao2ov3zpGpqTkyOn03nL1yCh4bKqP/v27SuXy6Vjx455xpSWliovL08dO3astfrQAFm9QogdpaenG5fLZdatW2cOHz5sJk+ebIKDg825c+eMMcaMGzfOzJs3zzM+MzPTNG7c2CxdutQcOXLEpKSkVLpkd3BwsNm8ebPZv3+/+cUvfnHLkt3Dhw83sbGxZteuXeaLL74wXbp0MQkJCfeucNQbVvXo119/bbKzs82iRYtMs2bNTHZ2tsnOzjaFhYX3rnj4PCv685tvvjFRUVFmyJAh5ptvvjFnz571PICbWdGfO3fuNCtWrDDZ2dkmLy/PbNu2zQwYMMB07tzZ/PDDD/f2DYDPs+p3/M2++uorVuNFpazozx07dpjly5ebnJwck5uba95//33TqlUrM378+HtbPHyeVdfP5557zrRr185s3brVHD161CQmJprWrVubK1eu3LviYTuEfXVkxYoVpkOHDqZp06bmgQceMDt37vTse+SRR8yECRO8xm/cuNF07drVNG3a1PTo0cNs2bLFa395eblJTk42YWFhxuVymSFDhphjx455jbl8+bJJSEgwzZo1M0FBQeaZZ54hRMFtWdGjEyZMMJJueXz22Wd1VSbqqXvdn2vXrq20N/mbGCpzr/tz//79ZvDgwaZFixbG5XKZiIgIM3XqVPPNN9/UaZ2ov6z4HX8zwj5U5V735969e018fLxxu93Gz8/PREdHm1dffZU/lqBSVlw/S0pKzKxZs0zr1q1N8+bNzdChQ83BgwfrrEY0DA5j+GZnAAAAAAAAwA74QiIAAAAAAADAJgj7AAAAAAAAAJsg7AMAAAAAAABsgrAPAAAAAAAAsAnCPgAAAAAAAMAmCPsAAAAAAAAAmyDsAwAAAAAAAGyCsA8AAAAAAACwCcI+AAAAAAAAwCYI+wAAAAAAAACbIOwDAAAAAAAAbOJ/Nfwn7Y0Iw4AAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# La 'feature importance' en modelos basados en árboles de decisión indican cuales son las características más importantes al momento de tomar una decisión\n",
    "pd.Series(search.best_estimator_.named_steps['classifier'].feature_importances_, index = search.best_estimator_.named_steps['vectorizer'].get_feature_names()).sort_values(ascending = False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of trees: 100\n",
      "Trees depth (mean): 180.21\n"
     ]
    }
   ],
   "source": [
    "estimators = search.best_estimator_.named_steps['classifier'].estimators_\n",
    "print('Number of trees:', len(estimators))\n",
    "print('Trees depth (mean):', np.mean([tree.get_depth() for tree in estimators]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x433978a30>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ConfusionMatrixDisplay.from_predictions(y_test, y_test_search_predict, normalize = 'true').plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x433af0d00>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ConfusionMatrixDisplay.from_predictions(y_test, y_test_search_predict, normalize = 'true').plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 1.0\n",
      "Recall: 1.0\n",
      "F1: 1.0\n"
     ]
    }
   ],
   "source": [
    "print('Precision:', precision_score(y_test.replace({'negative': 0, 'positive': 1}), y_test_search_predict))\n",
    "print('Recall:', recall_score(y_test.replace({'negative': 0, 'positive': 1}), y_test_search_predict))\n",
    "print('F1:', f1_score(y_test.replace({'negative': 0, 'positive': 1}), y_test_search_predict))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Support Vector Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to /Users/zejiran/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     /Users/zejiran/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package wordnet to /Users/zejiran/nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "import contractions\n",
    "import re\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.svm import SVC # import the SVC model\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.metrics import ConfusionMatrixDisplay, precision_score, recall_score, f1_score\n",
    "\n",
    "# download NLTK packages\n",
    "nltk.download('punkt')\n",
    "nltk.download('stopwords')\n",
    "nltk.download('wordnet')\n",
    "\n",
    "# read the data\n",
    "data=pd.read_csv('data/MovieReviews.csv', sep=',', encoding = 'utf-8')\n",
    "data_t=data\n",
    "data_t = data_t.drop(data_t.columns[0], axis=1)\n",
    "\n",
    "# define functions for preprocessing\n",
    "stop_words = set(stopwords.words('spanish'))\n",
    "def remove_non_ascii(words):\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = word.encode('utf-8', 'ignore').decode('utf-8', 'ignore')\n",
    "        new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def to_lowercase(words):\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = word.lower()\n",
    "        new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def remove_punctuation(words):\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = re.sub(r'[^\\w\\s]', '', word)\n",
    "        if new_word != '':\n",
    "            new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def replace_numbers(words):\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        new_word = re.sub(r'\\d+', '', word)\n",
    "        if new_word != '':\n",
    "            new_words.append(new_word)\n",
    "    return new_words\n",
    "\n",
    "def remove_stopwords(words):\n",
    "    \"\"\"Remove stop words from list of tokenized words\"\"\"\n",
    "    new_words = []\n",
    "    for word in words:\n",
    "        if word not in stop_words:\n",
    "            new_words.append(word)\n",
    "    return new_words\n",
    "\n",
    "def preprocessing(words):\n",
    "    words = to_lowercase(words)\n",
    "    words = replace_numbers(words)\n",
    "    words = remove_punctuation(words)\n",
    "    words = remove_non_ascii(words)\n",
    "    words = remove_stopwords(words)\n",
    "    return words\n",
    "\n",
    "# apply preprocessing to the data\n",
    "data_t['review_es'] = data_t['review_es'].apply(contractions.fix)\n",
    "data_t['words'] = data_t['review_es'].apply(word_tokenize)\n",
    "data_t['words'] = data_t['words'].apply(preprocessing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(data_t['review_es'], data_t['sentimiento'], test_size=0.2, random_state=42)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# create a pipeline that includes the SVC model\n",
    "pipeline = Pipeline([\n",
    "    ('tfidf', TfidfVectorizer(tokenizer=word_tokenize, stop_words=None)),\n",
    "    ('svc', SVC())\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# define a dictionary of parameters for the GridSearchCV\n",
    "parameters = {\n",
    "    'svc__C': [0.1, 1, 10],\n",
    "    'svc__kernel': ['linear', 'rbf']\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 6 candidates, totalling 30 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n",
      "/opt/homebrew/lib/python3.9/site-packages/sklearn/feature_extraction/text.py:528: UserWarning: The parameter 'token_pattern' will not be used since 'tokenizer' is not None'\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;tfidf&#x27;,\n",
       "                                        TfidfVectorizer(tokenizer=&lt;function word_tokenize at 0x1369538b0&gt;)),\n",
       "                                       (&#x27;svc&#x27;, SVC())]),\n",
       "             param_grid={&#x27;svc__C&#x27;: [0.1, 1, 10],\n",
       "                         &#x27;svc__kernel&#x27;: [&#x27;linear&#x27;, &#x27;rbf&#x27;]},\n",
       "             verbose=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 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 sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;tfidf&#x27;,\n",
       "                                        TfidfVectorizer(tokenizer=&lt;function word_tokenize at 0x1369538b0&gt;)),\n",
       "                                       (&#x27;svc&#x27;, SVC())]),\n",
       "             param_grid={&#x27;svc__C&#x27;: [0.1, 1, 10],\n",
       "                         &#x27;svc__kernel&#x27;: [&#x27;linear&#x27;, &#x27;rbf&#x27;]},\n",
       "             verbose=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 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 sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;tfidf&#x27;,\n",
       "                 TfidfVectorizer(tokenizer=&lt;function word_tokenize at 0x1369538b0&gt;)),\n",
       "                (&#x27;svc&#x27;, SVC())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">TfidfVectorizer</label><div class=\"sk-toggleable__content\"><pre>TfidfVectorizer(tokenizer=&lt;function word_tokenize at 0x1369538b0&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[('tfidf',\n",
       "                                        TfidfVectorizer(tokenizer=<function word_tokenize at 0x1369538b0>)),\n",
       "                                       ('svc', SVC())]),\n",
       "             param_grid={'svc__C': [0.1, 1, 10],\n",
       "                         'svc__kernel': ['linear', 'rbf']},\n",
       "             verbose=1)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# perform a grid search to find the best parameters\n",
    "grid_search = GridSearchCV(pipeline, parameters, cv=5, verbose=1)\n",
    "grid_search.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# evaluate the model on the test set\n",
    "y_pred = grid_search.predict(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# calculate precision, recall, and F1 score\n",
    "precision = precision_score(y_test, y_pred, average='weighted')\n",
    "recall = recall_score(y_test, y_pred, average='weighted')\n",
    "f1 = f1_score(y_test, y_pred, average='weighted')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.831829573934837\n",
      "Recall: 0.831\n",
      "F1 Score: 0.8308943089430896\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# print the performance metrics\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot the confusion matrix\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=['Negativo', 'Neutro', 'Positivo'])\n",
    "ax = plt.gca()\n",
    "disp.plot(include_values=True, cmap='Blues', ax=ax)\n",
    "ax.set_xticks([0, 1, 2, 3])\n",
    "ax.set_xticklabels(['Negativo', 'Neutro', 'Positivo'])\n",
    "plt.show()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clasificación con XGBoost y SentenceTransformers"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Implementado por: Daniel Reales"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Estrategia:\n",
    "\n",
    "- Vamos a generar embeddings del texto utilizando sentence transformers y posteriormente vamos a utilizarlos para entrenar un modelo de clasificación.\n",
    "- Podriamos generar embeddings para cada palabra y luego promediar los vectores. Sin embargo, podemos hacerlo directamente con un modelo preentrenado."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting xgboost\n",
      "  Downloading xgboost-1.7.5-py3-none-win_amd64.whl (70.9 MB)\n",
      "     ---------------------------------------- 70.9/70.9 MB 3.6 MB/s eta 0:00:00\n",
      "Requirement already satisfied: scipy in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from xgboost) (1.8.0)\n",
      "Requirement already satisfied: numpy in c:\\users\\fesar\\appdata\\local\\packages\\pythonsoftwarefoundation.python.3.10_qbz5n2kfra8p0\\localcache\\local-packages\\python310\\site-packages (from xgboost) (1.22.3)\n",
      "Installing collected packages: xgboost\n",
      "Successfully installed xgboost-1.7.5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 23.1\n",
      "[notice] To update, run: C:\\Users\\fesar\\AppData\\Local\\Microsoft\\WindowsApps\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "^C\n"
     ]
    }
   ],
   "source": [
    "%pip install xgboost\n",
    "%pip install sentence-transformers\n",
    "%pip install numpy --upgrade"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb\n",
    "import pandas as pd\n",
    "\n",
    "xgb_classifier = xgb.XGBClassifier()\n",
    "\n",
    "data = pd.read_csv('data/MovieReviews.csv', sep=',', encoding = 'utf-8')\n",
    "data = data.drop(data.columns[0], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "05f1b789c09a4c42972e043015829df0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)16ebc/.gitattributes:   0%|          | 0.00/737 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dbfced0a993f4c7986390bb0428f635e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)_Pooling/config.json:   0%|          | 0.00/190 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "383d5ff097b243bcab261aa6c14d5a5a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)b6b5d16ebc/README.md:   0%|          | 0.00/8.65k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bcb9a6b0e61f4fb8956198f56f405b8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)b5d16ebc/config.json:   0%|          | 0.00/571 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5fc51476129541ae87284ef5152fed5e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)ce_transformers.json:   0%|          | 0.00/116 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed045729bba24057a4c5cc08e62ff694",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)ebc/data_config.json:   0%|          | 0.00/25.5k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d47f54c2bbe34f71b5cfb34bcf0d431c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading pytorch_model.bin:   0%|          | 0.00/438M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ef1928e199ea429096a446eca5ff4851",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)nce_bert_config.json:   0%|          | 0.00/53.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "90948073219b4c01a6d3a27c717d06f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)cial_tokens_map.json:   0%|          | 0.00/239 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "041538a621894ef899e3c84e28b72b3a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)16ebc/tokenizer.json:   0%|          | 0.00/466k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "855bfcb4ce4f41aeb993b8a873752c96",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)okenizer_config.json:   0%|          | 0.00/363 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "243772c7bcd3454aaa2215a7e15fec81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)6ebc/train_script.py:   0%|          | 0.00/13.9k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "755e6f4a1da0420da7f89cd4ccf184f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)b6b5d16ebc/vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5ee5207ac6784c9097f9c2a662240420",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading (…)5d16ebc/modules.json:   0%|          | 0.00/229 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\fesar\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\torch\\_utils.py:147: UserWarning: Failed to initialize NumPy: module compiled against API version 0x10 but this version of numpy is 0xf . Check the section C-API incompatibility at the Troubleshooting ImportError section at https://numpy.org/devdocs/user/troubleshooting-importerror.html#c-api-incompatibility for indications on how to solve this problem . (Triggered internally at ..\\torch\\csrc\\utils\\tensor_numpy.cpp:84.)\n",
      "  t = torch.tensor([], dtype=storage.dtype, device=storage._untyped_storage.device)\n"
     ]
    }
   ],
   "source": [
    "from sentence_transformers import SentenceTransformer\n",
    "model = SentenceTransformer('multi-qa-mpnet-base-dot-v1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['sentimiento'] = data['sentimiento'].replace({'positivo': 0, 'negativo': 1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Splitear la data\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(data['review_es'], data['sentimiento'], test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "ename": "RuntimeError",
     "evalue": "Numpy is not available",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
      "\u001b[1;32mc:\\Users\\fesar\\git\\project-01\\movie_sentiment.ipynb Cell 61\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> <a href='vscode-notebook-cell:/c%3A/Users/fesar/git/project-01/movie_sentiment.ipynb#Y114sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m X_train_embeddings \u001b[39m=\u001b[39m model\u001b[39m.\u001b[39;49mencode(X_train\u001b[39m.\u001b[39;49mtolist())\n",
      "File \u001b[1;32m~\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sentence_transformers\\SentenceTransformer.py:197\u001b[0m, in \u001b[0;36mSentenceTransformer.encode\u001b[1;34m(self, sentences, batch_size, show_progress_bar, output_value, convert_to_numpy, convert_to_tensor, device, normalize_embeddings)\u001b[0m\n\u001b[0;32m    195\u001b[0m     all_embeddings \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mstack(all_embeddings)\n\u001b[0;32m    196\u001b[0m \u001b[39melif\u001b[39;00m convert_to_numpy:\n\u001b[1;32m--> 197\u001b[0m     all_embeddings \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39masarray([emb\u001b[39m.\u001b[39mnumpy() \u001b[39mfor\u001b[39;00m emb \u001b[39min\u001b[39;00m all_embeddings])\n\u001b[0;32m    199\u001b[0m \u001b[39mif\u001b[39;00m input_was_string:\n\u001b[0;32m    200\u001b[0m     all_embeddings \u001b[39m=\u001b[39m all_embeddings[\u001b[39m0\u001b[39m]\n",
      "File \u001b[1;32m~\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sentence_transformers\\SentenceTransformer.py:197\u001b[0m, in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m    195\u001b[0m     all_embeddings \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mstack(all_embeddings)\n\u001b[0;32m    196\u001b[0m \u001b[39melif\u001b[39;00m convert_to_numpy:\n\u001b[1;32m--> 197\u001b[0m     all_embeddings \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39masarray([emb\u001b[39m.\u001b[39;49mnumpy() \u001b[39mfor\u001b[39;00m emb \u001b[39min\u001b[39;00m all_embeddings])\n\u001b[0;32m    199\u001b[0m \u001b[39mif\u001b[39;00m input_was_string:\n\u001b[0;32m    200\u001b[0m     all_embeddings \u001b[39m=\u001b[39m all_embeddings[\u001b[39m0\u001b[39m]\n",
      "\u001b[1;31mRuntimeError\u001b[0m: Numpy is not available"
     ]
    }
   ],
   "source": [
    "X_train_embeddings = model.encode(X_train.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_embeddings = model.encode(X_test.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Fit the model\n",
    "xgb_classifier.fit(X_train_embeddings, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Predicciones\n",
    "y_pred = xgb_classifier.predict(X_test_embeddings)\n",
    "\n",
    "#Precision, recall y f1\n",
    "from sklearn.metrics import precision_score, recall_score, f1_score\n",
    "precision = precision_score(y_test, y_pred, average='weighted')\n",
    "recall = recall_score(y_test, y_pred, average='weighted')\n",
    "f1 = f1_score(y_test, y_pred, average='weighted')\n",
    "\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Matriz de confusion\n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "#Plot de la matriz de confusion\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=['Negativo', 'Positivo'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.11"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "25e464d3af9e3fc5bb7c31aad8b4ecf6ce2814819f704e4acaa7662244112331"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}