{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Pandas\n",
    "\n",
    "[Pandas](http://pandas.pydata.org/) je biblioteka za analizu podataka. Ilustrirat ćemo njeno korištenje na jednostavnom primjeru analize podataka iz IMDB baze.\n",
    "\n",
    "Koristit ću i biblioteke [requests](https://github.com/kennethreitz/requests) (za učitavanje web stranica) i [BeautifulSoup](https://www.crummy.com/software/BeautifulSoup/) (za analizu HTML-a). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import requests\n",
    "from bs4 import BeautifulSoup as bs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Dohvaćamo podatke s IMDB-a o filmovima sa zemljom porijekla Hrvatska, smimljenim između 1945. i 2017."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import time\n",
    "url = 'http://www.imdb.com/search/title'\n",
    "params = dict(sort='num_votes,desc', start=1, title_type='feature', year='1945,2017',countries='hr', languages='hr')\n",
    "r=[]\n",
    "for i in range(1,240,50): # takvih filmova ima trenutno 201\n",
    "    params['start']=i\n",
    "    r.append(requests.get(url, params=params))\n",
    "    time.sleep(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Parsiranje podataka koje spremamo u datoteku `filmovi2.txt`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import re\n",
    "with open('filmovi2.txt','w') as f:\n",
    "    for i in range(len(r)):\n",
    "        soup = bs(r[i].text,'html.parser')\n",
    "        for film in soup.find_all('div', class_=\"lister-item\"):\n",
    "            for a in film.find_all('a', href=True):\n",
    "                if '/title/tt'  and 'adv_li_tt' in a['href']:\n",
    "                    title = a.contents[0]\n",
    "            rt = film.find_all('span', class_=\"runtime\")\n",
    "            if rt:\n",
    "                runtime =  rt[0].contents[0]\n",
    "            else:\n",
    "                runtime = '0 mins'\n",
    "            y = film.find_all('span', class_=\"lister-item-year\")\n",
    "            if y: \n",
    "                year = y[0].contents[0]\n",
    "                year = year.replace('(I)','').replace('(III)','').strip()\n",
    "            else:\n",
    "                year = '???'\n",
    "            rat = film.find_all('span', class_='global-sprite rating-star imdb-rating')\n",
    "            if rat:\n",
    "                rating = str(list(rat[0].next_siblings)[1]).replace('<strong>','').replace('</strong>','')\n",
    "            else:\n",
    "                rat = '0'\n",
    "            g = film.find_all('span',class_=\"genre\")\n",
    "            if g: genres =  ' '.join(g[0].contents).replace('\\n','').strip()\n",
    "            d=film.find_all(string =re.compile('Director'))[0]\n",
    "            director = d.next_element.contents[0]\n",
    "            f.write('\\t'.join((title, year, runtime, rating,director, genres))+'\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Hunting Party\t(2007)\t101 min\t6.9\tRichard Shepard\tAdventure, Comedy, Drama\r\n",
      "Karaula\t(2006)\t94 min\t7.7\tRajko Grlic\tAction, Comedy, Drama\r\n",
      "Svecenikova djeca\t(2013)\t96 min\t6.8\tVinko Bresan\tComedy, Drama\r\n",
      "Metastaze\t(2009)\t82 min\t7.8\tBranko Schmidt\tCrime, Drama\r\n",
      "Zvizdan\t(2015)\t123 min\t7.3\tDalibor Matanic\tDrama, Romance, War\r\n",
      "How the War Started on My Island\t(1996)\t97 min\t7.9\tVinko Bresan\tComedy, War\r\n",
      "Ljudozder vegetarijanac\t(2012)\t85 min\t7.2\tBranko Schmidt\tDrama\r\n",
      "Fine mrtve djevojke\t(2002)\t77 min\t7.2\tDalibor Matanic\tDrama, Thriller\r\n",
      "Sonja i bik\t(2012)\t103 min\t7.1\tVlatka Vorkapic\tComedy, Romance\r\n",
      "Petelinji zajtrk\t(2007)\t124 min\t7.6\tMarko Nabersnik\tDrama\r\n"
     ]
    }
   ],
   "source": [
    "!head filmovi2.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Analiza podataka pomoću biblioteke Pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows: 201\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>runtime</th>\n",
       "      <th>rating</th>\n",
       "      <th>director</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Hunting Party</td>\n",
       "      <td>(2007)</td>\n",
       "      <td>101 min</td>\n",
       "      <td>6.9</td>\n",
       "      <td>Richard Shepard</td>\n",
       "      <td>Adventure, Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Karaula</td>\n",
       "      <td>(2006)</td>\n",
       "      <td>94 min</td>\n",
       "      <td>7.7</td>\n",
       "      <td>Rajko Grlic</td>\n",
       "      <td>Action, Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Svecenikova djeca</td>\n",
       "      <td>(2013)</td>\n",
       "      <td>96 min</td>\n",
       "      <td>6.8</td>\n",
       "      <td>Vinko Bresan</td>\n",
       "      <td>Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Metastaze</td>\n",
       "      <td>(2009)</td>\n",
       "      <td>82 min</td>\n",
       "      <td>7.8</td>\n",
       "      <td>Branko Schmidt</td>\n",
       "      <td>Crime, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Zvizdan</td>\n",
       "      <td>(2015)</td>\n",
       "      <td>123 min</td>\n",
       "      <td>7.3</td>\n",
       "      <td>Dalibor Matanic</td>\n",
       "      <td>Drama, Romance, War</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               title    year  runtime  rating         director  \\\n",
       "0  The Hunting Party  (2007)  101 min     6.9  Richard Shepard   \n",
       "1            Karaula  (2006)   94 min     7.7      Rajko Grlic   \n",
       "2  Svecenikova djeca  (2013)   96 min     6.8     Vinko Bresan   \n",
       "3          Metastaze  (2009)   82 min     7.8   Branko Schmidt   \n",
       "4            Zvizdan  (2015)  123 min     7.3  Dalibor Matanic   \n",
       "\n",
       "                     genres  \n",
       "0  Adventure, Comedy, Drama  \n",
       "1     Action, Comedy, Drama  \n",
       "2             Comedy, Drama  \n",
       "3              Crime, Drama  \n",
       "4       Drama, Romance, War  "
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "names = ['title', 'year','runtime', 'rating', 'director', 'genres']\n",
    "data = pd.read_csv('filmovi2.txt', delimiter='\\t', names=names)\n",
    "print (\"Number of rows: {:d}\".format(data.shape[0]))\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>runtime</th>\n",
       "      <th>rating</th>\n",
       "      <th>director</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Hunting Party</td>\n",
       "      <td>2007</td>\n",
       "      <td>101</td>\n",
       "      <td>6.9</td>\n",
       "      <td>Richard Shepard</td>\n",
       "      <td>Adventure, Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Karaula</td>\n",
       "      <td>2006</td>\n",
       "      <td>94</td>\n",
       "      <td>7.7</td>\n",
       "      <td>Rajko Grlic</td>\n",
       "      <td>Action, Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Svecenikova djeca</td>\n",
       "      <td>2013</td>\n",
       "      <td>96</td>\n",
       "      <td>6.8</td>\n",
       "      <td>Vinko Bresan</td>\n",
       "      <td>Comedy, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Metastaze</td>\n",
       "      <td>2009</td>\n",
       "      <td>82</td>\n",
       "      <td>7.8</td>\n",
       "      <td>Branko Schmidt</td>\n",
       "      <td>Crime, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Zvizdan</td>\n",
       "      <td>2015</td>\n",
       "      <td>123</td>\n",
       "      <td>7.3</td>\n",
       "      <td>Dalibor Matanic</td>\n",
       "      <td>Drama, Romance, War</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               title  year  runtime  rating         director  \\\n",
       "0  The Hunting Party  2007      101     6.9  Richard Shepard   \n",
       "1            Karaula  2006       94     7.7      Rajko Grlic   \n",
       "2  Svecenikova djeca  2013       96     6.8     Vinko Bresan   \n",
       "3          Metastaze  2009       82     7.8   Branko Schmidt   \n",
       "4            Zvizdan  2015      123     7.3  Dalibor Matanic   \n",
       "\n",
       "                     genres  \n",
       "0  Adventure, Comedy, Drama  \n",
       "1     Action, Comedy, Drama  \n",
       "2             Comedy, Drama  \n",
       "3              Crime, Drama  \n",
       "4       Drama, Romance, War  "
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data['runtime'].fillna('0 mins.', inplace=True);\n",
    "clean_runtime = [int(v.split(' ')[0]) for v in data.runtime]\n",
    "data['runtime'] = clean_runtime\n",
    "data['year'] = [int(y[1:-1]) for y in data.year]\n",
    "# data.rating[data.rating=='-'] = '0';\n",
    "clean_rating = [float(v) for v in data.rating]\n",
    "data['rating'] = clean_rating\n",
    "#clean_genres = [g.replace(' ','|') for g in data.genres]\n",
    "#data['genres'] = clean_genres\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "title       Doktor ludosti\n",
       "year                  2003\n",
       "runtime                  0\n",
       "rating                   7\n",
       "director      Fadil Hadzic\n",
       "genres              Comedy\n",
       "Name: 118, dtype: object"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.ix[118]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>runtime</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>201.000000</td>\n",
       "      <td>201.000000</td>\n",
       "      <td>201.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2007.935323</td>\n",
       "      <td>85.029851</td>\n",
       "      <td>7.131841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.379078</td>\n",
       "      <td>29.859489</td>\n",
       "      <td>0.866534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1970.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2004.000000</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>6.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2010.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>7.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2013.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>7.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2017.000000</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>9.600000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              year     runtime      rating\n",
       "count   201.000000  201.000000  201.000000\n",
       "mean   2007.935323   85.029851    7.131841\n",
       "std       7.379078   29.859489    0.866534\n",
       "min    1970.000000    0.000000    3.700000\n",
       "25%    2004.000000   78.000000    6.700000\n",
       "50%    2010.000000   90.000000    7.200000\n",
       "75%    2013.000000  100.000000    7.600000\n",
       "max    2017.000000  200.000000    9.600000"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[['year','runtime', 'rating']].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "data.replace(0,np.nan, inplace=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>runtime</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>186.000000</td>\n",
       "      <td>201.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>91.887097</td>\n",
       "      <td>7.131841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>18.176407</td>\n",
       "      <td>0.866534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>46.000000</td>\n",
       "      <td>3.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>80.250000</td>\n",
       "      <td>6.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>91.500000</td>\n",
       "      <td>7.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>7.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>200.000000</td>\n",
       "      <td>9.600000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          runtime      rating\n",
       "count  186.000000  201.000000\n",
       "mean    91.887097    7.131841\n",
       "std     18.176407    0.866534\n",
       "min     46.000000    3.700000\n",
       "25%     80.250000    6.700000\n",
       "50%     91.500000    7.200000\n",
       "75%    100.000000    7.600000\n",
       "max    200.000000    9.600000"
      ]
     },
     "execution_count": 218,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[['runtime', 'rating']].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEKCAYAAAACS67iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEctJREFUeJzt3XmwpFV9xvHvI2gQwQjOQHDBwRSoqIg6KnEFLQlCGSDG\nimgiLuVYFqY0apJJTCpWTKowMSZxiQYVwbhvRFKgQgg6rsgM6yAqi6MhTjkYVHApEPzlj/cMNJd7\n5259tzPfT1VXv336XX73pXnm7dN9TqeqkCStfHdb6gIkSeNhoEtSJwx0SeqEgS5JnTDQJakTBrok\ndcJAl6ROGOiS1AkDXZI6setiHmzVqlW1Zs2axTykJK14mzZt+mFVrZ5uvUUN9DVr1rBx48bFPKQk\nrXhJvjuT9exykaROGOiS1AkDXZI6YaBLUicMdEnqhIEuSZ0w0CWpEwa6JHXCQJekTizqSFFJ2hms\nWX/WXdq2nHzMgh/XK3RJ6oSBLkmdMNAlqRMGuiR1wkCXpE4Y6JLUCQNdkjphoEtSJwx0SeqEgS5J\nnTDQJakTBrokdcJAl6ROGOiS1IlpAz3JA5Ocn+TKJFckeVVr3zvJuUmuavd7LXy5kqSpzOQK/Vbg\ntVX1MOAw4KQkBwPrgfOq6kDgvPZYkrREpg30qtpaVRe15ZuAK4H7A8cCp7fVTgeOW6giJUnTm1Uf\nepI1wKOBC4B9q2orDKEP7DPu4iRJMzfjQE+yB/BJ4NVVdeMstluXZGOSjddff/1capQkzcCMAj3J\n3RnC/INV9anW/IMk+7Xn9wO2TbZtVZ1SVWurau3q1avHUbMkaRIz+ZZLgPcCV1bVW0aeOhM4sS2f\nCHx6/OVJkmZq1xms8yTgD4HLk1zS2v4COBn4WJKXAt8DnrswJUqSZmLaQK+qLwGZ4ulnjLccSdJc\nOVJUkjphoEtSJwx0SeqEgS5JnTDQJakTBrokdcJAl6ROGOiS1AkDXZI6YaBLUicMdEnqhIEuSZ0w\n0CWpEzOZPleSNGLN+rPu9HjLyccsUSV35hW6JHXCQJekThjoktQJA12SOmGgS1InDHRJ6oSBLkmd\nMNAlqRMGuiR1wkCXpE4Y6JLUCQNdkjphoEtSJwx0SeqEgS5JnTDQJakTBrokdcJAl6ROGOiS1AkD\nXZI6YaBLUicMdEnqhIEuSZ0w0CWpE9MGepJTk2xLsnmk7Q1J/jfJJe129MKWKUmazkyu0E8Djpqk\n/Z+q6tB2O3u8ZUmSZmvaQK+qDcANi1CLJGke5tOH/sokl7Uumb3GVpEkaU52neN27wTeCFS7/0fg\nJZOtmGQdsA5g//33n+PhJGn5WrP+rKUuAZjjFXpV/aCqbquqXwHvBh6/g3VPqaq1VbV29erVc61T\nkjSNOQV6kv1GHh4PbJ5qXUnS4pi2yyXJh4HDgVVJrgP+Gjg8yaEMXS5bgJcvYI2SpBmYNtCr6oRJ\nmt+7ALVIkubBkaKS1AkDXZI6YaBLUicMdEnqxFwHFklSF5bLoKBx8ApdkjphoEtSJwx0SeqEgS5J\nnTDQJakTBrokdcJAl6ROGOiS1AkHFknaqfQ0kGgir9AlqRMGuiR1wkCXpE4Y6JLUCQNdkjphoEtS\nJwx0SeqEgS5JnTDQJakTBrokdcJAl6ROGOiS1AkDXZI6YaBLUicMdEnqhIEuSZ0w0CWpEwa6JHXC\nQJekThjoktQJA12SOmGgS1InDHRJ6oSBLkmdmDbQk5yaZFuSzSNteyc5N8lV7X6vhS1TkjSdmVyh\nnwYcNaFtPXBeVR0InNceS5KW0LSBXlUbgBsmNB8LnN6WTweOG3NdkqRZmmsf+r5VtRWg3e8zvpIk\nSXOx4B+KJlmXZGOSjddff/1CH06SdlpzDfQfJNkPoN1vm2rFqjqlqtZW1drVq1fP8XCSpOnMNdDP\nBE5syycCnx5POZKkuZrJ1xY/DHwVeEiS65K8FDgZeGaSq4BntseSpCW063QrVNUJUzz1jDHXIkma\nB0eKSlInDHRJ6oSBLkmdMNAlqRMGuiR1wkCXpE4Y6JLUCQNdkjphoEtSJwx0SeqEgS5JnTDQJakT\nBrokdWLa2RYlaaVYs/6sOz3ecvIxS1TJ0vAKXZI6YaBLUicMdEnqhIEuSZ0w0CWpEwa6JHXCQJek\nThjoktQJA12SOuFIUUnzNo4RmhP3MRM720jQ6XiFLkmdMNAlqRMGuiR1wkCXpE4Y6JLUCQNdkjph\noEtSJwx0SeqEA4sk7dDO/rNuK4lX6JLUCQNdkjphoEtSJwx0SeqEgS5JnZjXt1ySbAFuAm4Dbq2q\nteMoSpI0e+P42uIRVfXDMexHkjQPdrlIUifmG+gFnJNkU5J14yhIkjQ38+1yeVJVfT/JPsC5Sb5Z\nVRtGV2hBvw5g//33n+fhJK1Ek/283DhGnE73s3Vz+Vm7lWxeV+hV9f12vw04A3j8JOucUlVrq2rt\n6tWr53M4SdIOzDnQk9wryZ7bl4Ejgc3jKkySNDvz6XLZFzgjyfb9fKiqPjuWqiRJszbnQK+qa4FH\njbEWSdI8+LVFSeqEgS5JnTDQJakTBrokdcKfoJM0KzvbYJ2VxCt0SeqEgS5JnTDQJakTBrokdcJA\nl6ROGOiS1AkDXZI6YaBLUicMdEnqhCNFpRVqqX7WbVz7cMTp+HmFLkmdMNAlqRMGuiR1wkCXpE4Y\n6JLUCQNdkjphoEtSJwx0SeqEA4ukjkwcrDOTgUYO8OmHV+iS1AkDXZI6YaBLUicMdEnqhIEuSZ0w\n0CWpEwa6JHXCQJekTjiwSGO3UL+ksxDmMhBnJXHQ0M7FK3RJ6oSBLkmdMNAlqRMGuiR1wkCXpE7M\nK9CTHJXkW0muTrJ+XEVJkmZvzoGeZBfgHcCzgIOBE5IcPK7CJEmzM58r9McDV1fVtVV1C/AR4Njx\nlCVJmq35BPr9gf8ZeXxda5MkLYH5jBTNJG11l5WSdcC69vCnSb41j2MuhFXAD5e6iBlYKXXCJLXm\nTUtUyfTuVOsyrhNW+GtgGVuUWuf52nrQTFaaT6BfBzxw5PEDgO9PXKmqTgFOmcdxFlSSjVW1dqnr\nmM5KqROsdaFY68JYSbVOZz5dLhcCByY5IMk9gOcBZ46nLEnSbM35Cr2qbk3ySuBzwC7AqVV1xdgq\nkyTNyrxmW6yqs4Gzx1TLUlm23UETrJQ6wVoXirUujJVU6w6l6i6fY0qSViCH/ktSJ7oL9CSnJtmW\nZPNI26OSfDXJ5Un+M8m9W/uaJL9Ickm7vWtkm8e29a9O8tYkk31Nc9Fqbc8d0p67oj2/23KsNckL\nRs7pJUl+leTQZVrr3ZOc3tqvTPLnI9ss+NQWs6z1Hkne19ovTXL4yDYLel6TPDDJ+e0cXZHkVa19\n7yTnJrmq3e/V2tPquDrJZUkeM7KvE9v6VyU5cZx1zrHWh7bzfXOS103Y18qa3qSquroBTwUeA2we\nabsQeFpbfgnwxra8ZnS9Cfv5OvBbDN+3/wzwrCWudVfgMuBR7fF9gV2WY60TtnskcO0yPq/PBz7S\nlncHtrTXxS7ANcCDgXsAlwIHL3GtJwHva8v7AJuAuy3GeQX2Ax7TlvcEvs0w5cffA+tb+3rgTW35\n6FZHgMOAC1r73sC17X6vtrzXEte6D/A44O+A143sZ1FeA+O8dXeFXlUbgBsmND8E2NCWzwWes6N9\nJNkPuHdVfbWG/7LvB45b4lqPBC6rqkvbtv9XVbct01pHnQB8GJbteS3gXkl2Be4J3ALcyCJNbTHL\nWg8GzmvbbQN+DKxdjPNaVVur6qK2fBNwJcPI8GOB09tqp48c91jg/TX4GnCfVudvA+dW1Q1V9aP2\n9x21lLVW1baquhD45YRdrbjpTboL9ClsBn6nLT+XOw+IOiDJxUm+kOQpre3+DAOntlvMaQ2mqvUg\noJJ8LslFSf60tS/HWkf9Pi3QWZ61fgL4GbAV+B7w5qq6gaWd2mKqWi8Fjk2ya5IDgMe25xb1vCZZ\nAzwauADYt6q2whCkDFe7MPX5W9TzOsNap7LipjfZWQL9JcBJSTYxvAW7pbVvBfavqkcDrwE+1Por\nZzStwQKZqtZdgScDL2j3xyd5BsuzVgCSPAH4eVVt7x9ejrU+HrgNuB9wAPDaJA9medZ6KkOobAT+\nGfgKcCuLWGuSPYBPAq+uqht3tOoUNS3HWqfcxSRty/prgfP6HvpKUVXfZOiyIMlBwDGt/Wbg5ra8\nKck1DFfC1zFMZbDdpNMaLGatraYvVNUP23NnM/S9fmAZ1rrd87jj6hyW53l9PvDZqvolsC3Jl4G1\nDFdm005tsZi1VtWtwB9vXy/JV4CrgB+xCOc1yd0ZAvKDVfWp1vyDJPtV1dbWpbKttU81Nch1wOET\n2j+/xLVOZUbTmywnO8UVepJ92v3dgL8E3tUer84wrzvtquxAhg/wtgI3JTmsfVvghcCnl7JWhhG5\nhyTZvfX3Pg34xjKtdXvbcxn6HYHb3+Yut1q/Bzy9fSvjXgwf4H2TJZzaYgev191bjSR5JnBrVS3K\na6Dt973AlVX1lpGnzgS2f1PlxJHjngm8sJ3Xw4CftDo/BxyZZK/2LZMjW9tS1jqVlTe9yVJ/Kjvu\nG8MV4VaGDziuA14KvIrhk+5vAydzx4Cq5wBXMPRNXgQ8e2Q/axn6Mq8B3r59m6Wqta3/B63ezcDf\nL/NaDwe+Nsl+llWtwB7Ax9t5/QbwJyP7Obqtfw3w+mXwel0DfIvhQ77/Ah60WOeVoZuvGL5pdUm7\nHc3wbavzGN4pnAfs3dYPww/gXANcDqwd2ddLgKvb7cULcE5nW+tvtHN/I8MHzdcxfMi8KK+Bcd4c\nKSpJndgpulwkaWdgoEtSJwx0SeqEgS5JnTDQJakTBroWRJJ9k3woybVJNrXZ7I6f5T4+n2RtWz47\nyX0Wptq5S/KGiTP0TbLOi5K8fRb7/Eq7v1+ST8y3Ru08DHSNXRvY8R/Ahqp6cFU9lmFQxgN2vOXU\nquroqvrxuGrcke2DzZZKVT2x3X+/qn5vKWvRymKgayE8Hbilqm4fOVpV362qtwEk2S13zOt9cZIj\nWvs9k3wkwzzWZzDMfkh7bkuSVRnmsL8yybszzHV9TpJ7tnVeluTCDHOFfzLJ7hMLa1fU/97eMVyV\n5GWt/fAkX0xyJsPAHZK8Jsnmdnv1yD5en+TbSb7EMDPi9vbRdxSrkmyZ5PjHtGOvau9izmj1Xprk\niW2dn7b7NWnzpCfZJck/tL/vsiQvn+N/G3Vsp5jLRYvu4Qwjb6dyEkBVPTLJQ4FzMsxZ8gqGybwe\nluSQHezjQOCEqnpZko8xjPj9APCpqno3QJK/ZRh1+bZJtj+EYYj/vYCLk5zV2h8DPKKqvpPkscCL\ngScwjHq8IMkXGC6CngccyvD/z0UM85JPq3U5vQY4uqp+lOSjDPPzHN/eFeyxg81fyjB8/nFJfg34\ncpJzquo7Mzm2dg4GuhZckncwDMe+paoe15bfBsNEVEm+yzAp2lOBt7b2y5JcNsUuv1NVl7TlTQxD\n4gEe0YL8PgzhONUcIZ+uql8Av0hyPsOMiz8Gvj4SkE8Gzqiqn7W/4VPAUxgC/Yyq+nlrn+ncHkcw\nDM8/su6Y+e/pDPOuUFW3AT/ZwfZHMszls70L5tcZ/mEz0HU7A10L4QpGfuyiqk5Ksoph2leYfFrS\n21efwf5vHlm+jTu6Zk4DjquqS5O8iDvP6rejY2x//LORtrnUeCt3dGPuNuG5axl++eYg7jgPsxHg\nj6pqrBNZqS/2oWsh/DewW5JXjLSN9mdvYJjXffv0sPszTDo12v4Ihq6R2dgT2Jph6tQX7GC9Y1s/\n/n0ZQv/CSdbZABw3MsPh8cAXW/vxrb9/T+DZI9tsYfjRCYCJH2Z+F/hd4P1JHt7azmPoZtreR35v\npvY54BXtbyPJQa0u6XYGusauhhnfjgOeluQ7Sb7O8JNff9ZW+VdglySXAx8FXlTD3PTvBPZIciXw\nN8ywb3rEXzH8Ms2XGabAncplwPnA1xh+r/Muc1zX8BNmpzH8VucFwHuq6uLW/lGGGTo/w53/MXgz\nQ+heDKyaZJ/fYviH5uNJfpNhVsUj2nnYxPDZw102a/fvYZgN8qL2Qem/4TtsTeBsi9qpJHkD8NOq\nevNS1zKd9sHsW6rqaUtdi1YGr9ClZah9/fHDwL8sdS1aObxCl6ROeIUuSZ0w0CWpEwa6JHXCQJek\nThjoktQJA12SOvH/tCT9pYEINCoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc6fccea6a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.hist(data.year, bins=np.arange(1945, 2017))\n",
    "plt.xlabel(\"Godina produkcije\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEKCAYAAAACS67iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqr06GbxBzwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc6fccc39e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(data.rating.dropna(), bins=20)\n",
    "plt.xlabel(\"IMDB ocjena\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEKCAYAAAARnO4WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6/oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc7041ba978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(data.year, data.rating, lw=0, color='k')\n",
    "plt.xlabel(\"Godina\")\n",
    "plt.ylabel(\"IMDB ocjena\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>rating</th>\n",
       "      <th>director</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>Larin izbor: Izgubljeni princ</td>\n",
       "      <td>2012</td>\n",
       "      <td>3.7</td>\n",
       "      <td>Tomislav Rukavina</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            title  year  rating           director genres\n",
       "83  Larin izbor: Izgubljeni princ  2012     3.7  Tomislav Rukavina  Drama"
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.rating == data.rating.min()][['title', 'year', 'rating','director', 'genres']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>rating</th>\n",
       "      <th>director</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>The Second Death of Maximilian Paspa Aka Paspi...</td>\n",
       "      <td>2016</td>\n",
       "      <td>9.6</td>\n",
       "      <td>Zorko Sirotic</td>\n",
       "      <td>Mystery</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 title  year  rating  \\\n",
       "189  The Second Death of Maximilian Paspa Aka Paspi...  2016     9.6   \n",
       "\n",
       "          director   genres  \n",
       "189  Zorko Sirotic  Mystery  "
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.rating == data.rating.max()][['title', 'year', 'rating','director',  'genres']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>runtime</th>\n",
       "      <th>rating</th>\n",
       "      <th>director</th>\n",
       "      <th>genres</th>\n",
       "      <th>Adventure</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Drama</th>\n",
       "      <th>...</th>\n",
       "      <th>Biography</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Family</th>\n",
       "      <th>Fantasy</th>\n",
       "      <th>Horror</th>\n",
       "      <th>Mystery</th>\n",
       "      <th>Thriller</th>\n",
       "      <th>War</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Hunting Party</td>\n",
       "      <td>2007</td>\n",
       "      <td>101.0</td>\n",
       "      <td>6.9</td>\n",
       "      <td>Richard Shepard</td>\n",
       "      <td>Adventure, Comedy, Drama</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Karaula</td>\n",
       "      <td>2006</td>\n",
       "      <td>94.0</td>\n",
       "      <td>7.7</td>\n",
       "      <td>Rajko Grlic</td>\n",
       "      <td>Action, Comedy, Drama</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Svecenikova djeca</td>\n",
       "      <td>2013</td>\n",
       "      <td>96.0</td>\n",
       "      <td>6.8</td>\n",
       "      <td>Vinko Bresan</td>\n",
       "      <td>Comedy, Drama</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Metastaze</td>\n",
       "      <td>2009</td>\n",
       "      <td>82.0</td>\n",
       "      <td>7.8</td>\n",
       "      <td>Branko Schmidt</td>\n",
       "      <td>Crime, Drama</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Zvizdan</td>\n",
       "      <td>2015</td>\n",
       "      <td>123.0</td>\n",
       "      <td>7.3</td>\n",
       "      <td>Dalibor Matanic</td>\n",
       "      <td>Drama, Romance, War</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               title  year  runtime  rating         director  \\\n",
       "0  The Hunting Party  2007    101.0     6.9  Richard Shepard   \n",
       "1            Karaula  2006     94.0     7.7      Rajko Grlic   \n",
       "2  Svecenikova djeca  2013     96.0     6.8     Vinko Bresan   \n",
       "3          Metastaze  2009     82.0     7.8   Branko Schmidt   \n",
       "4            Zvizdan  2015    123.0     7.3  Dalibor Matanic   \n",
       "\n",
       "                     genres  Adventure  Comedy  Crime  Drama  ...   Biography  \\\n",
       "0  Adventure, Comedy, Drama      False    True  False   True  ...       False   \n",
       "1     Action, Comedy, Drama      False    True  False   True  ...       False   \n",
       "2             Comedy, Drama      False   False  False   True  ...       False   \n",
       "3              Crime, Drama      False   False  False   True  ...       False   \n",
       "4       Drama, Romance, War      False   False  False  False  ...       False   \n",
       "\n",
       "  Comedy  Crime  Drama Family Fantasy Horror Mystery Thriller    War  \n",
       "0  False  False  False  False   False  False   False    False  False  \n",
       "1  False  False  False  False   False  False   False    False  False  \n",
       "2   True  False  False  False   False  False   False    False  False  \n",
       "3  False   True  False  False   False  False   False    False  False  \n",
       "4  False  False   True  False   False  False   False    False  False  \n",
       "\n",
       "[5 rows x 35 columns]"
      ]
     },
     "execution_count": 225,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "genres = set()\n",
    "for m in data.genres:\n",
    "    genres.update(g for g in m.split(','))\n",
    "genres = sorted(genres)\n",
    "\n",
    "for genre in genres:\n",
    "    data[genre] = [genre in movie.split(',') for movie in data.genres]\n",
    "         \n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Genre Count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Adventure</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Comedy</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Crime</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drama</th>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Family</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fantasy</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>History</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Horror</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Music</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Musical</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mystery</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Romance</th>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sci-Fi</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sport</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thriller</th>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>War</th>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Action</th>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adventure</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Animation</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Biography</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Comedy</th>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Crime</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drama</th>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Family</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fantasy</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Horror</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mystery</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thriller</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>War</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Genre Count\n",
       " Adventure            2\n",
       " Comedy               7\n",
       " Crime                4\n",
       " Drama               43\n",
       " Family               9\n",
       " Fantasy              2\n",
       " History              5\n",
       " Horror               2\n",
       " Music                2\n",
       " Musical              1\n",
       " Mystery              5\n",
       " Romance             16\n",
       " Sci-Fi               2\n",
       " Sport                1\n",
       " Thriller            12\n",
       " War                 19\n",
       "Action               15\n",
       "Adventure             6\n",
       "Animation             3\n",
       "Biography             6\n",
       "Comedy               49\n",
       "Crime                10\n",
       "Drama                97\n",
       "Family                3\n",
       "Fantasy               2\n",
       "Horror                3\n",
       "Mystery               4\n",
       "Thriller              2\n",
       "War                   1"
      ]
     },
     "execution_count": 226,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "genre_count = data[genres].sum()\n",
    "pd.DataFrame({'Genre Count': genre_count})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>year</th>\n",
       "      <th>petoljetka</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Hunting Party</td>\n",
       "      <td>2007</td>\n",
       "      <td>2005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Karaula</td>\n",
       "      <td>2006</td>\n",
       "      <td>2005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Svecenikova djeca</td>\n",
       "      <td>2013</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Metastaze</td>\n",
       "      <td>2009</td>\n",
       "      <td>2005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Zvizdan</td>\n",
       "      <td>2015</td>\n",
       "      <td>2015</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               title  year  petoljetka\n",
       "0  The Hunting Party  2007        2005\n",
       "1            Karaula  2006        2005\n",
       "2  Svecenikova djeca  2013        2010\n",
       "3          Metastaze  2009        2005\n",
       "4            Zvizdan  2015        2015"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "petoljetka =  (data.year // 5) * 5\n",
    "\n",
    "tyd = data.loc[:, ('title', 'year')]\n",
    "tyd['petoljetka'] = petoljetka;\n",
    "\n",
    "tyd.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "year\n",
      "1970    8.600000\n",
      "1980    7.700000\n",
      "1990    7.114286\n",
      "1995    6.754545\n",
      "2000    6.956000\n",
      "2005    6.973684\n",
      "2010    7.193333\n",
      "2015    7.512500\n",
      "Name: Petoljetka mean, dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEKCAYAAAARnO4WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxpQIC/jGGFMi/j+W8g9XIHQScQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc6fce2e0b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pet_mean = data.groupby(petoljetka).rating.mean()\n",
    "pet_mean.name = 'Petoljetka mean'\n",
    "print (pet_mean)\n",
    "\n",
    "plt.plot(pet_mean.index, pet_mean.values, 'o-',\n",
    "        color='r', lw=3, label='Petoljetka prosjek')\n",
    "plt.scatter(data.year, data.rating, alpha=.04, lw=0, color='k')\n",
    "plt.xlabel(\"Godina\")\n",
    "plt.ylabel(\"Ocjena\")\n",
    "plt.legend(frameon=False);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1970 ['Tko pjeva zlo ne misli']\n",
      "1980 ['Izgubljeni zavicaj']\n",
      "1991 ['Vrijeme ratnika']\n",
      "1992 ['Kamenita vrata']\n",
      "1993 ['Kontesa Dora']\n",
      "1994 ['Vukovar se vraca kuci']\n",
      "1995 ['Washed Out']\n",
      "1996 ['How the War Started on My Island']\n",
      "1997 ['Pont Neuf']\n",
      "1998 ['Kad mrtvi zapjevaju']\n",
      "1999 ['Marsal' 'Dubrovacki suton']\n",
      "2000 ['Je li jasno prijatelju?']\n",
      "2001 ['Holding']\n",
      "2002 ['24 sata' 'Serafin, svjetionicarev sin']\n",
      "2003 ['Konjanik' 'Tu' 'Bore Lee: U kandzama velegrada']\n",
      "2004 ['Nije bed']\n",
      "2005 ['RGB: RedGreenBlue']\n",
      "2006 ['Crveno i crno']\n",
      "2007 ['Pjevajte nesto ljubavno']\n",
      "2008 ['Blazeni Augustin Kazotic']\n",
      "2009 ['Da mogu...']\n",
      "2010 ['Mjesto na kojem je umro posljednji covjek']\n",
      "2011 [\"Marija's Own\"]\n",
      "2012 [\"Once Upon a Winter's Night\"]\n",
      "2013 ['Glazbena kutija']\n",
      "2014 ['Vlog']\n",
      "2015 ['Oporuka']\n",
      "2016 ['The Second Death of Maximilian Paspa Aka Paspin Kut3']\n",
      "2017 ['Uzbuna na Zelenom Vrhu']\n"
     ]
    }
   ],
   "source": [
    "for year, subset in data.groupby('year'):\n",
    "    print (year, subset[subset.rating == subset.rating.max()].title.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table><tr><td><strong>Python verzija</strong></td><td>3.5.3</td></tr><tr><td><strong>kompajler</strong></td><td>GCC 4.8.2 20140120 (Red Hat 4.8.2-15)</td></tr><tr><td><strong>sustav</strong></td><td>Linux</td></tr><tr><td><strong>broj CPU-a</strong></td><td>8</td></tr><tr><td><strong>interpreter</strong></td><td>64bit</td></tr></table><table><tr><td><strong>pandas</strong> verzija</td><td>0.19.2</td></tr><tr><td><strong> numpy</strong> verzija</td><td>1.11.3</td></tr><tr><td><strong>requests</strong> verzija</td><td>2.13.0</td></tr><tr><td><strong> beautifulsoup4</strong> verzija</td><td>4.5.3</td></tr></table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 230,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from verzije import *\n",
    "from IPython.display import HTML\n",
    "HTML(print_sysinfo()+info_packages('pandas, numpy,requests, beautifulsoup4'))"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 3 (Anaconda)",
   "language": "python",
   "name": "anaconda3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.3"
  },
  "name": "pandas.ipynb"
 },
 "nbformat": 4,
 "nbformat_minor": 0
}