main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from utils import normalize\n",
    "from kmeans import KMeansModel\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ArielMartinezSilberstein/Library/Python/3.7/lib/python/site-packages/IPython/core/interactiveshell.py:2785: DtypeWarning: Columns (10,23,33,37) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>eleccion</th>\n",
       "      <th>vueltae</th>\n",
       "      <th>cargo</th>\n",
       "      <th>tipocargo</th>\n",
       "      <th>fechae</th>\n",
       "      <th>anoe</th>\n",
       "      <th>periodo</th>\n",
       "      <th>fechainicio</th>\n",
       "      <th>diainicio</th>\n",
       "      <th>mesinicio</th>\n",
       "      <th>...</th>\n",
       "      <th>siglapart_text</th>\n",
       "      <th>pacto_text</th>\n",
       "      <th>subpacto_text</th>\n",
       "      <th>aux_folio</th>\n",
       "      <th>aux_partido</th>\n",
       "      <th>aux_pacto</th>\n",
       "      <th>aux_subpacto</th>\n",
       "      <th>vototot</th>\n",
       "      <th>voto_h</th>\n",
       "      <th>voto_m</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Presidenciales</td>\n",
       "      <td>Primera</td>\n",
       "      <td>Presidente</td>\n",
       "      <td>Eleccion Popular</td>\n",
       "      <td>14-12-89</td>\n",
       "      <td>1989</td>\n",
       "      <td>1990-1994</td>\n",
       "      <td>11-03-90</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>IND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IND NA NA</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>NaN</td>\n",
       "      <td>962</td>\n",
       "      <td>505.0</td>\n",
       "      <td>457.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Presidenciales</td>\n",
       "      <td>Primera</td>\n",
       "      <td>Presidente</td>\n",
       "      <td>Eleccion Popular</td>\n",
       "      <td>14-12-89</td>\n",
       "      <td>1989</td>\n",
       "      <td>1990-1994</td>\n",
       "      <td>11-03-90</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>IND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IND NA NA</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>NaN</td>\n",
       "      <td>369</td>\n",
       "      <td>190.0</td>\n",
       "      <td>179.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Presidenciales</td>\n",
       "      <td>Primera</td>\n",
       "      <td>Presidente</td>\n",
       "      <td>Eleccion Popular</td>\n",
       "      <td>14-12-89</td>\n",
       "      <td>1989</td>\n",
       "      <td>1990-1994</td>\n",
       "      <td>11-03-90</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>IND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IND NA NA</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>NaN</td>\n",
       "      <td>203</td>\n",
       "      <td>106.0</td>\n",
       "      <td>97.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Presidenciales</td>\n",
       "      <td>Primera</td>\n",
       "      <td>Presidente</td>\n",
       "      <td>Eleccion Popular</td>\n",
       "      <td>14-12-89</td>\n",
       "      <td>1989</td>\n",
       "      <td>1990-1994</td>\n",
       "      <td>11-03-90</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>IND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IND NA NA</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>NaN</td>\n",
       "      <td>219</td>\n",
       "      <td>117.0</td>\n",
       "      <td>102.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Presidenciales</td>\n",
       "      <td>Primera</td>\n",
       "      <td>Presidente</td>\n",
       "      <td>Eleccion Popular</td>\n",
       "      <td>14-12-89</td>\n",
       "      <td>1989</td>\n",
       "      <td>1990-1994</td>\n",
       "      <td>11-03-90</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>IND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IND NA NA</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>IndFP</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2660</td>\n",
       "      <td>1358.0</td>\n",
       "      <td>1302.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         eleccion  vueltae       cargo         tipocargo    fechae  anoe  \\\n",
       "0  Presidenciales  Primera  Presidente  Eleccion Popular  14-12-89  1989   \n",
       "1  Presidenciales  Primera  Presidente  Eleccion Popular  14-12-89  1989   \n",
       "2  Presidenciales  Primera  Presidente  Eleccion Popular  14-12-89  1989   \n",
       "3  Presidenciales  Primera  Presidente  Eleccion Popular  14-12-89  1989   \n",
       "4  Presidenciales  Primera  Presidente  Eleccion Popular  14-12-89  1989   \n",
       "\n",
       "     periodo fechainicio  diainicio  mesinicio  ... siglapart_text pacto_text  \\\n",
       "0  1990-1994    11-03-90         11          3  ...            IND        NaN   \n",
       "1  1990-1994    11-03-90         11          3  ...            IND        NaN   \n",
       "2  1990-1994    11-03-90         11          3  ...            IND        NaN   \n",
       "3  1990-1994    11-03-90         11          3  ...            IND        NaN   \n",
       "4  1990-1994    11-03-90         11          3  ...            IND        NaN   \n",
       "\n",
       "   subpacto_text  aux_folio  aux_partido  aux_pacto aux_subpacto  vototot  \\\n",
       "0            NaN  IND NA NA        IndFP      IndFP          NaN      962   \n",
       "1            NaN  IND NA NA        IndFP      IndFP          NaN      369   \n",
       "2            NaN  IND NA NA        IndFP      IndFP          NaN      203   \n",
       "3            NaN  IND NA NA        IndFP      IndFP          NaN      219   \n",
       "4            NaN  IND NA NA        IndFP      IndFP          NaN     2660   \n",
       "\n",
       "   voto_h  voto_m  \n",
       "0   505.0   457.0  \n",
       "1   190.0   179.0  \n",
       "2   106.0    97.0  \n",
       "3   117.0   102.0  \n",
       "4  1358.0  1302.0  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"data.csv\", sep=\";\", decimal=\",\")\n",
    "data = data.loc[:, ~data.columns.str.contains('^Unnamed')]\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>voto_m</th>\n",
       "      <th>voto_h</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>457.0</td>\n",
       "      <td>505.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>179.0</td>\n",
       "      <td>190.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>97.0</td>\n",
       "      <td>106.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>102.0</td>\n",
       "      <td>117.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1302.0</td>\n",
       "      <td>1358.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   voto_m  voto_h\n",
       "0   457.0   505.0\n",
       "1   179.0   190.0\n",
       "2    97.0   106.0\n",
       "3   102.0   117.0\n",
       "4  1302.0  1358.0"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_name, y_name = \"voto_m\", \"voto_h\"\n",
    "source = data[np.logical_and(np.isfinite(data[x_name]), np.isfinite(data[y_name]))]\n",
    "source = source[[x_name, y_name]]\n",
    "source.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>voto_m</th>\n",
       "      <th>voto_h</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.007210</td>\n",
       "      <td>0.010003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.002824</td>\n",
       "      <td>0.003763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.001530</td>\n",
       "      <td>0.002100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.001609</td>\n",
       "      <td>0.002317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.020541</td>\n",
       "      <td>0.026899</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     voto_m    voto_h\n",
       "0  0.007210  0.010003\n",
       "1  0.002824  0.003763\n",
       "2  0.001530  0.002100\n",
       "3  0.001609  0.002317\n",
       "4  0.020541  0.026899"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "source = normalize(source, source.min(), source.max())\n",
    "source.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = KMeansModel()\n",
    "centroids, clusters = model.cluster(source, 3, 6)\n",
    "\n",
    "result = []\n",
    "for idx, cluster in clusters.items():\n",
    "    for vector in cluster:\n",
    "        result.append({\n",
    "            x_name: vector[0],\n",
    "            y_name: vector[1],\n",
    "            \"label\": idx\n",
    "        })\n",
    "result = pd.DataFrame(result)\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 426.475x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rc(\"font\", size=16)\n",
    "sns.set_style(\"white\")\n",
    "palette = [\"#630C3A\", \"#39C8C6\", \"#D3500C\"]\n",
    "sns.set_palette(palette)\n",
    "facet = sns.lmplot(data=result, x=x_name, y=y_name, hue=\"label\", fit_reg=False, legend=True, legend_out=True)"
   ]
  },
  {
   "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}