Years.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from engine.models.movie import Movie\n",
    "\n",
    "DATASET = 'dataset/expanded_movies.txt'\n",
    "folder = '/media/mariam/Files/ran/clacket-save'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calc_years(interval=None):\n",
    "    result = {}\n",
    "    with open(DATASET) as dataset:\n",
    "        for movie_line in dataset:\n",
    "            year = movie_line.split('|')[2]\n",
    "            if year != 'N/A':\n",
    "                year = int(year)\n",
    "                if year not in result:\n",
    "                    result[year] = 0\n",
    "                result[year] += 1\n",
    "    if interval is not None:\n",
    "        prev = []\n",
    "        final = {}\n",
    "        for year in result:\n",
    "            if len(prev) == interval:\n",
    "                last = prev[-1]['year']\n",
    "                final[last] = sum([p['val'] for p in prev])\n",
    "                prev = []\n",
    "            prev.append({'year': year, 'val': result[year]})\n",
    "        last = prev[-1]['year']\n",
    "        final[last] = sum([p['val'] for p in prev])\n",
    "        return pd.Series(final)\n",
    "    else:\n",
    "        return pd.Series(result)\n",
    "    \n",
    "def calc_ratings(interval=None):\n",
    "    result = {}\n",
    "    with open(DATASET) as dataset:\n",
    "        for movie_line in dataset:\n",
    "            spl = movie_line.split('|')\n",
    "            movie_id, year = spl[0], spl[2]\n",
    "            if year != 'N/A':\n",
    "                movie = Movie.load(folder, movie_id)\n",
    "                ratings = movie.ratings.size\n",
    "                if year not in result:\n",
    "                    result[year] = 0\n",
    "                result[year] += ratings\n",
    "    if interval is not None:\n",
    "        prev = []\n",
    "        final = {}\n",
    "        for year in result:\n",
    "            if len(prev) == interval:\n",
    "                last = prev[-1]['year']\n",
    "                final[last] = sum([p['val'] for p in prev])\n",
    "                prev = []\n",
    "            prev.append({'year': year, 'val': result[year]})\n",
    "        last = prev[-1]['year']\n",
    "        final[last] = sum([p['val'] for p in prev])\n",
    "        return pd.Series(final)\n",
    "    else:\n",
    "        return pd.Series(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = calc_years(interval=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = calc_ratings(interval=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pylab\n",
    "\n",
    "def plot(df, kind='line', start=1980, end=2006, step=10, prefix=''):\n",
    "    ran = range(start, end)\n",
    "    #df_ = df.filter(items=ran)\n",
    "    # df.plot(kind=kind, xticks=range(start, end, step))\n",
    "    df.plot(kind=kind, x='type', ylim=(0,1.5))\n",
    "    plt.title('Variation of MAE for k=2, Random')\n",
    "    saved = pylab.savefig('Results_mae_types_ran.png'.format(kind, start, end, prefix))\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7185b54e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7180e99588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(df, start=1970)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7180e460b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(df, start=1910)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEYCAYAAAB82RxTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ncEdEtEwKd0REy6RwR0S0zP8HNWbqYx2JPW0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7182f024a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(df2, kind='bar', start=1910, prefix='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "res = pd.read_pickle('80_20_min_max.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "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>type</th>\n",
       "      <th>MAE_3</th>\n",
       "      <th>MAE_4</th>\n",
       "      <th>MAE_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mean</td>\n",
       "      <td>0.446</td>\n",
       "      <td>0.443</td>\n",
       "      <td>0.505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>max</td>\n",
       "      <td>0.744</td>\n",
       "      <td>0.790</td>\n",
       "      <td>0.998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>min</td>\n",
       "      <td>0.670</td>\n",
       "      <td>0.681</td>\n",
       "      <td>0.842</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   type  MAE_3  MAE_4  MAE_5\n",
       "0  mean  0.446  0.443  0.505\n",
       "1   max  0.744  0.790  0.998\n",
       "2   min  0.670  0.681  0.842"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEpCAYAAABoRGJ5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVkAOdzOzAnK4m5kVkMPdzKyA/j+aTp0tNckrBAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f71806fddd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(res, kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_pickle('80_20.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>k</th>\n",
       "      <th>mean_absolute_error_3</th>\n",
       "      <th>mean_absolute_error_4</th>\n",
       "      <th>mean_absolute_error_5</th>\n",
       "      <th>perc_hits_3</th>\n",
       "      <th>perc_hits_4</th>\n",
       "      <th>perc_hits_5</th>\n",
       "      <th>perc_misses_3</th>\n",
       "      <th>perc_misses_4</th>\n",
       "      <th>perc_misses_5</th>\n",
       "      <th>rmse_3</th>\n",
       "      <th>rmse_4</th>\n",
       "      <th>rmse_5</th>\n",
       "      <th>total_hits_3</th>\n",
       "      <th>total_hits_4</th>\n",
       "      <th>total_hits_5</th>\n",
       "      <th>total_misses_3</th>\n",
       "      <th>total_misses_4</th>\n",
       "      <th>total_misses_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.250936</td>\n",
       "      <td>0.242842</td>\n",
       "      <td>0.271889</td>\n",
       "      <td>96.559781</td>\n",
       "      <td>89.930340</td>\n",
       "      <td>83.253134</td>\n",
       "      <td>3.440219</td>\n",
       "      <td>10.069660</td>\n",
       "      <td>16.746866</td>\n",
       "      <td>0.700751</td>\n",
       "      <td>0.700751</td>\n",
       "      <td>0.700751</td>\n",
       "      <td>1605485</td>\n",
       "      <td>967985</td>\n",
       "      <td>356753</td>\n",
       "      <td>57200</td>\n",
       "      <td>108387</td>\n",
       "      <td>71763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.445686</td>\n",
       "      <td>0.443283</td>\n",
       "      <td>0.505147</td>\n",
       "      <td>93.706914</td>\n",
       "      <td>81.044680</td>\n",
       "      <td>69.248031</td>\n",
       "      <td>6.293086</td>\n",
       "      <td>18.955320</td>\n",
       "      <td>30.751969</td>\n",
       "      <td>0.918352</td>\n",
       "      <td>0.918352</td>\n",
       "      <td>0.918352</td>\n",
       "      <td>2363279</td>\n",
       "      <td>1307027</td>\n",
       "      <td>420294</td>\n",
       "      <td>158711</td>\n",
       "      <td>305697</td>\n",
       "      <td>186646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0.546595</td>\n",
       "      <td>0.546344</td>\n",
       "      <td>0.632274</td>\n",
       "      <td>92.398497</td>\n",
       "      <td>76.842884</td>\n",
       "      <td>61.368781</td>\n",
       "      <td>7.601503</td>\n",
       "      <td>23.157116</td>\n",
       "      <td>38.631219</td>\n",
       "      <td>1.006340</td>\n",
       "      <td>1.006340</td>\n",
       "      <td>1.006340</td>\n",
       "      <td>3090556</td>\n",
       "      <td>1633108</td>\n",
       "      <td>474431</td>\n",
       "      <td>254256</td>\n",
       "      <td>492148</td>\n",
       "      <td>298651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.602603</td>\n",
       "      <td>0.591459</td>\n",
       "      <td>0.676588</td>\n",
       "      <td>91.731032</td>\n",
       "      <td>75.332081</td>\n",
       "      <td>58.620780</td>\n",
       "      <td>8.268968</td>\n",
       "      <td>24.667919</td>\n",
       "      <td>41.379220</td>\n",
       "      <td>1.056594</td>\n",
       "      <td>1.056594</td>\n",
       "      <td>1.056594</td>\n",
       "      <td>3694715</td>\n",
       "      <td>1876434</td>\n",
       "      <td>513732</td>\n",
       "      <td>333055</td>\n",
       "      <td>614449</td>\n",
       "      <td>362633</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   k  mean_absolute_error_3  mean_absolute_error_4  mean_absolute_error_5  \\\n",
       "0  1               0.250936               0.242842               0.271889   \n",
       "1  2               0.445686               0.443283               0.505147   \n",
       "2  3               0.546595               0.546344               0.632274   \n",
       "3  4               0.602603               0.591459               0.676588   \n",
       "\n",
       "   perc_hits_3  perc_hits_4  perc_hits_5  perc_misses_3  perc_misses_4  \\\n",
       "0    96.559781    89.930340    83.253134       3.440219      10.069660   \n",
       "1    93.706914    81.044680    69.248031       6.293086      18.955320   \n",
       "2    92.398497    76.842884    61.368781       7.601503      23.157116   \n",
       "3    91.731032    75.332081    58.620780       8.268968      24.667919   \n",
       "\n",
       "   perc_misses_5    rmse_3    rmse_4    rmse_5  total_hits_3  total_hits_4  \\\n",
       "0      16.746866  0.700751  0.700751  0.700751       1605485        967985   \n",
       "1      30.751969  0.918352  0.918352  0.918352       2363279       1307027   \n",
       "2      38.631219  1.006340  1.006340  1.006340       3090556       1633108   \n",
       "3      41.379220  1.056594  1.056594  1.056594       3694715       1876434   \n",
       "\n",
       "   total_hits_5  total_misses_3  total_misses_4  total_misses_5  \n",
       "0        356753           57200          108387           71763  \n",
       "1        420294          158711          305697          186646  \n",
       "2        474431          254256          492148          298651  \n",
       "3        513732          333055          614449          362633  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for min in [3,4,5]:\n",
    "    df['total_{0}'.format(min)] = df['total_hits_{0}'.format(min)] + df['total_misses_{0}'.format(min)]"
   ]
  },
  {
   "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}