analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### imports etc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>type</th>\n",
       "      <th>address</th>\n",
       "      <th>url</th>\n",
       "      <th>postcode</th>\n",
       "      <th>number_bedrooms</th>\n",
       "      <th>search_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3012.0</td>\n",
       "      <td>2 bedroom flat</td>\n",
       "      <td>Horsley Court, Montaigne Close, London</td>\n",
       "      <td>/property-to-rent/property-44782458.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1100.0</td>\n",
       "      <td>1 bedroom flat</td>\n",
       "      <td>Beaconsfield Road, London</td>\n",
       "      <td>/property-to-rent/property-44818353.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1450.0</td>\n",
       "      <td>2 bedroom flat</td>\n",
       "      <td>Gaumont House,  Marmont Road,  London</td>\n",
       "      <td>/property-to-rent/property-44818389.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1257.0</td>\n",
       "      <td>3 bedroom flat</td>\n",
       "      <td>Lady Hamilton Court,  Holders Hill Avenue,  Lo...</td>\n",
       "      <td>/property-to-rent/property-44818368.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1300.0</td>\n",
       "      <td>1 bedroom flat</td>\n",
       "      <td>Elm Park, London, SW2</td>\n",
       "      <td>/property-to-rent/property-56484454.html</td>\n",
       "      <td>SW2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    price            type                                            address  \\\n",
       "0  3012.0  2 bedroom flat             Horsley Court, Montaigne Close, London   \n",
       "1  1100.0  1 bedroom flat                          Beaconsfield Road, London   \n",
       "2  1450.0  2 bedroom flat              Gaumont House,  Marmont Road,  London   \n",
       "3  1257.0  3 bedroom flat  Lady Hamilton Court,  Holders Hill Avenue,  Lo...   \n",
       "4  1300.0  1 bedroom flat                              Elm Park, London, SW2   \n",
       "\n",
       "                                        url postcode  number_bedrooms  \\\n",
       "0  /property-to-rent/property-44782458.html      NaN              2.0   \n",
       "1  /property-to-rent/property-44818353.html      NaN              1.0   \n",
       "2  /property-to-rent/property-44818389.html      NaN              2.0   \n",
       "3  /property-to-rent/property-44818368.html      NaN              3.0   \n",
       "4  /property-to-rent/property-56484454.html      SW2              1.0   \n",
       "\n",
       "  search_date  \n",
       "0  11/10/2016  \n",
       "1  11/10/2016  \n",
       "2  11/10/2016  \n",
       "3  11/10/2016  \n",
       "4  11/10/2016  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "import matplotlib.mlab as mlab\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# Read in the csv file for analysis\n",
    "df = pd.read_csv('london_rent.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>number_bedrooms</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4968.000000</td>\n",
       "      <td>4938.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2394.276369</td>\n",
       "      <td>1.924868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1911.129822</td>\n",
       "      <td>1.121821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>390.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1452.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1863.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2600.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>30333.000000</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              price  number_bedrooms\n",
       "count   4968.000000      4938.000000\n",
       "mean    2394.276369         1.924868\n",
       "std     1911.129822         1.121821\n",
       "min      390.000000         0.000000\n",
       "25%     1452.000000         1.000000\n",
       "50%     1863.000000         2.000000\n",
       "75%     2600.000000         2.000000\n",
       "max    30333.000000         8.000000"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quick look at the shape of the data.\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Flat share', 'Detached house', 'Property', 'Town house', 'House',\n",
       "       'Terraced house'], dtype=object)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# See which 'types' don't have bedroom number extracted\n",
    "df[df.number_bedrooms.isnull()].type.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Create a data frame with summary statistics by number of bedrooms.\n",
    "df_by_bedroom = pd.DataFrame(columns = ['number_bedrooms','count','average_price'])\n",
    "df_by_bedroom['number_bedrooms'] = df.number_bedrooms[df.number_bedrooms.notnull()].unique().astype(int)\n",
    "df_by_bedroom.sort_values(by='number_bedrooms',inplace=True)\n",
    "df_by_bedroom.reset_index(drop=True,inplace=True)\n",
    "for i in range(0,len(df_by_bedroom.index),1):\n",
    "    df_by_bedroom['count'].loc[i]=df[df.number_bedrooms==i].number_bedrooms.count()\n",
    "    df_by_bedroom['average_price'].loc[i]=df[df.number_bedrooms==i].price.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmgAAAH4CAYAAAD+YRGXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5uqcL8oAjAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1203ae410>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm8AAAH4CAYAAAAcvQruAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XuYZVV95//3R1ouKiBooASExiAKKkbU9j6UmkBIIjI6\nYscoKMT5jTBKMtEEyM+H1vxi4iU/0TFoVETwEgYZFTQEEaEyIYqgqKCgtJG70ka5xEtELt/5Y+/C\nQ1FVnO6uU6dW9fv1POfpfdZee++1zqnu+vRa+5KqQpIkSW14wLgbIEmSpOEZ3iRJkhpieJMkSWqI\n4U2SJKkhhjdJkqSGGN4kSZIaYniTNDZJvpnkPy3wPi9IcvhC7nPUkuyW5O4kY/k3Ockzk1yV5N+T\nHDTL+quTPG+BjnVykjcvxL6kTdWKcTdA0qarqh4/7jYsIeO86eabgXdX1XvG2AZJQ3LkTVqmkmw2\n7jbMZSm3rXUb+NnuBlyx0G1ZX/5cSMMxvEljkOTPkny3n6b6ZpKD+/LNk9ySZO+Bug9P8vMkD+/f\n/16Sr/X1LkzyhIG6Vyf50yTfAH6a5AFzHauv/4Akf5Pk35L8a5KjBqfvkmyT5INJvp/k+iR/kSRz\n9On4JJ9Iclp/rK8k2Weetm02OB3Xt+W4vq23Jbkkyc79uscmOTfJj5NcmeQl9/MR75Hky/1+PpXk\nof1+PpvkqBnt/kaSF87Sn+mpzEOTXJvkh0mOG1h/r+m/JPsluX5Gf1/f7/8nST6QZIckZ/efz7lJ\nth08JHBEkhv7158M7CtJjuk/m3/rP+PpPk238/Ak1wJfmOP7eXWStUl+lOTTSSb68u8CuwOf7dv1\nwDk+01VJvtV/Bycl2Xxg3/P9TD4pyVf77+I0YMuZn1n/c/ED4ENztPURA9s8M8nF/bG+nOQZA+su\n6H9G/6X/zM9Msn2Sj/bH/3KSXQfqvzPJun7dNzLw905a0qrKly9fi/wCXgzs2C+/BPjpwPsPAn8x\nUPdI4Ox++UnAOuApdL/sXwFcDTywX381cCmwE7DFEMf6b8A3gUcA2wKfB+4CHtCv/xRwIt0v3IcD\nFwGvnqNPxwO3A/8Z2Az4E+B7wGbztO1q4Hn98huAbwB79O+fAGwHPAi4Dji07/MTgR8Cj52jHRcA\n1wN7AVsBZwCnDvT/ooG6TwT+DVgxy352A+4G/g7YHNgH+AXwmH79ycCbB+rvB1w38P5q4Iv95/aI\n/nv7Sr+fzelC1htnHOtj/Wf9+L6P05/N0f2+HgE8EHgv8PEZ23647+8Ws/TleX0/n9hv/27gn2a0\n9bnz/LxeDVzWf3cPBS6c7jvz/Ez2r2uA1/U/Ey8Gfjmw7X7AHcBb+rpbzNfW/ufhZuBldIMPq/v3\n2w1891cBK4GtgW8B3wae29c/BTipr7s/cAmwdf/+MfR/L3z5WuqvsTfAly9fBfA14AX98vOB7w6s\nuxD4g375ROBNM7b9NvCcfvlq4LD1ONYXGAhj/bHv6n/R7UgXVrYYWL8aOH+O/R4PfHHgfYDvA8+a\nq23cO7x9G/i9WfZ7yGDQ6MveRx98Zql/AfCWgfd70YXK9OHgx8Cv9+veDrxnjv3s1n8Wjxgo+zJw\nSL88THj7/YH3ZwB/O/D+vwOfHDjW3cCjB9a/FfhAv3wFA+GKLsT9sv+eptu52zzf+QeBvx54/+B+\n+11nfg9zbH/1jJ+TA4G19/cz2b9umLHuX7h3ePsF/X8+5mnr7cCuwMsZCN/9+i8Chw5898cOrHsH\n8A8D738PuLRffm7fzqcBWai/y758LcbLaVNpDPqpuOlppluAx9GN0ED3C2irJE9NshvdCMSn+3W7\nAX+S5Ob+dQuwC92IyLQb1uNYO9GNUk0bXN6VbuTjBwPHet/AtrO5Z/uqqr4tc7ZthkfSjdTNtBvw\n9Bl9fhkwMUw7gGv7fjy8qm4HTgdeniTA7wMfmWc/0I0qTfs58JD7qT/Xtv8xy/vBfU1/XoPtnv7s\ndgM+Nf0Z0IW5O+gC9rT5Ptud+v11B6r6GV2I3Xm4btxn/zPbNtfP5E7AjTP2c+2M9/9WVXfcT1tv\n7tt6r3UD+xvsx1CfeVVdALwH+FtgXZL3JVmf71YaG8ObtMj6c27eDxxZVdtV1XZ00zsBqKq76QLG\ny+jCxWf7X2DQhZK/rKrt+9d2VfWQqvpfA4eoYY8F/IDuF+20XQeWr6cbFXnYwLEeWlX7MLdHDhw7\n/b4Hf3nPd0Xl9cCvz1E+NaPP21TVUbPUvU876MLFL4Ef9e9PoRvBeT7ws6r68jz7mc/P6KZ0pz1i\nrorrYbDdu9KNXEI3bXzgjM/gwVX1g4H6832236f7HABI8mDgYcwf+OZr224DbZvvZ/IH3Dcg7jrj\n/cx2z9XWG/t1K2fZ38yAOJSqek9VPQXYm27a9A0bsh9psRnepMX3YLopsh+lO0n/VXTnOA36e+Cl\ndAHu4wPlHwD+W5JV0P1iS/I7/S+4DTnW6cDRSXbqT4D/0+kVVXUTcC7wziRb9yfNPyrz35ftyUkO\nTnfV4B/Thb9hw9EHgb9Iskfftyck2Q74LLBnkpcnWZHkgUmekuSx8+zr5ekucngQ8CbgE/1IIFV1\nUf+Z/A33P+o268UZva8Dv5Nku/7k/6OH6uX8x3pjkq2SPA54FXBav+7vgLdMn2yf5Ndy7/uxzddO\n6H6eXpVknyRb0J1jdlFVXX8/2w06KsnOSbYHjhto23w/k18C7kzy2v67exGwagPbeh1wNvDoJKvT\nXfDyUrpp8c+sRz/o2/mUJKuSrKAbkfsF3c+FtOQZ3qRFVlVX0gWHi4Cb6KYxL5xR52K6kZ1HAP84\nUP5V4NXAe/rps6uAwwY3Xc9jfYAuoF0GfBX4B+DOfvQPuosENqebprsZ+ATzT1eeSRc6bwH+AHhR\nVd01W9tmKfv/6cLkuUluowtzW1XVT+lOLl9NN/LyfeCv+3bNpuhC2Sl93c25b7A6lS7EfnSevszW\n5sH3H6H73K4BzuFXYWaYbec61j8B36W7cORtVTV95ei76D7b6c/mi9w7BM27734/bwQ+STdKtTvd\n57k+bfs43c/Kd4G1wF/2+57zZ7KfDn0RXRD9Md0FI/97Q9taVTfTnbf2erqR1NcDv1tVtwzZj0Hb\n0P3830x3Tt+P6M6BlJa89P8ZHc3Ok5Po/qKtG5xqSfJauivo7qQ7mfSYvvxY4PC+/OiqOrcv35fu\nSqot6a66+6O+fHO6f4SfTPcX76X9/84kbYAkvw28t6p234Btj6e7EODQhW/ZwkryCroT8Bf06Q6S\ntBhGPfJ2MnDAYEGSSeAFwBOq6gl0VwORZC+6q8r2oruS6cT+nBnoLos/oqr2pJs+md7nEcDNVfVo\n4ATgbaPtjrS8JNkyyYH9FNTOdFeMfnLc7Rqlfir1SLqpSElqzkjDW1VdSDd9Mug1dJeB39nXmT6J\n+IXAaVV1Z1VdQzcsv6o/l2Trqrqkr3cqcPDANqf0y2fQnYAsaXihOyfsZrpp02/RBbhlKcn+dPdP\n+wHduVWS1JxxPNt0T+A/JXkL3Umir+/PmdiZ7uTWaTf2ZXdy7yuibuBXVy/tTH9LgKq6K8mtSbbv\nz4uQdD+q6j+4/xPIh93XmxZiP6PUn4rh7SAkNW0c4W0F3d2wn57kqXQnQD9qgfY95xVXSUZ3cp8k\nSdICq6pZc804rja9nv6cmn4q9K4k0/fwGbz/z/T9oW7k3vcXGrxv1D3r+lsTbDPfqNu474g8ytfx\nxx8/9jb48rvbFF9+f22//P7afS33724+ixHewr1HxD5N9+w6kuwJbF5VPwbOAl6a7sHcuwN7ABdX\nd6+p2/r78YTu1gVn9vs6i1/dJuElwPkj740kSdIYjXTaNMnHgUngYUmuozsR+kPAyUkup3te3aEA\nVXVFktP51WNfjqxfRc+juPetQs7py08CPpJkLd09hAbvWyRJkrTsjDS8VdXL5lj1ijnq/xXwV7OU\nfxV4wizlt9PdXmSTNzk5Oe4maAP53bXN769tfn/t2pS/u5HepHcpSVKbSl8lSVLbklBL6IIFSZIk\nbSDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElS\nQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkN\nMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE\n8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDD\nmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkjRGExMrSdLUa2Ji5bg/tk1a\nqmrcbVgUSWpT6askqR1JgNZ+PwV/p45WEqoqs61z5E2SJKkhIw1vSU5Ksi7JZbOs+5MkdyfZfqDs\n2CRrk1yZZP+B8n2TXJbkqiQnDJRvnuS0fpsvJdl1lP2RJEkat1GPvJ0MHDCzMMkuwG8B1w6U7QUc\nAuwFHAicmG4sGeC9wBFVtSewZ5LpfR4B3FxVjwZOAN42qo5IkiQtBSMNb1V1IXDLLKveCbxhRtkL\ngdOq6s6qugZYC6xKMgFsXVWX9PVOBQ4e2OaUfvkM4PkL2HxJkqQlZ9HPeUtyEHB9VV0+Y9XOwPUD\n72/sy3YGbhgov6Evu9c2VXUXcOvgNKwkSdJys2IxD5ZkK+A4uinTkRxivpVr1qy5Z3lycpLJyckR\nNUOSJGl4U1NTTE1NDVV35LcKSbIb8Jmq2ifJ44HzgJ/TBa1d6EbYVgGHA1TVX/fbnQMcT3de3AVV\ntVdfvhrYr6peM12nqr6cZDPgB1W1wxzt8FYhkqQlx1uFaDbjvlVI+hdV9c2qmqiqR1XV7nRToE+q\nqh8CZwEv7a8g3R3YA7i4qm4Cbkuyqr+A4VDgzH7fZwGH9csvAc5fhP5IkiSNzahvFfJx4It0V4he\nl+RVM6oUvwp2VwCnA1cAZwNHDgyVHQWcBFwFrK2qc/ryk4CHJ1kL/BFwzCj7I0mSNG4+YUGSpDFy\n2lSzGfe0qSRJkhaI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4\nkyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFN\nkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJ\nkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJ\nkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJ\naojhTZIkqSGGN0mSpIaMNLwlOSnJuiSXDZS9LcmVSb6e5H8n2WZg3bFJ1vbr9x8o3zfJZUmuSnLC\nQPnmSU7rt/lSkl1H2R9JkqRxG/XI28nAATPKzgUeV1W/AawFjgVIsjdwCLAXcCBwYpL027wXOKKq\n9gT2TDK9zyOAm6vq0cAJwNtG2RlJkqRxG2l4q6oLgVtmlJ1XVXf3by8CdumXDwJOq6o7q+oaumC3\nKskEsHVVXdLXOxU4uF9+IXBKv3wG8PyRdESSJGmJGPc5b4cDZ/fLOwPXD6y7sS/bGbhhoPyGvuxe\n21TVXcCtSbYfZYMlSZLGacW4Dpzkz4E7qurvF3K3861cs2bNPcuTk5NMTk4u4KElSZI2zNTUFFNT\nU0PVTVWNtDFJdgM+U1X7DJS9Eng18Lyqur0vOwaoqnpr//4c4HjgWuCCqtqrL18N7FdVr5muU1Vf\nTrIZ8IOq2mGOdtSo+ypJ0vrqTu9u7fdT8HfqaCWhqmYdlFqMadMwMCKW5LeBNwAHTQe33lnA6v4K\n0t2BPYCLq+om4LYkq/oLGA4FzhzY5rB++SXA+aPtiiRJ0niNdNo0yceBSeBhSa6jG0k7Dtgc+Hx/\nMelFVXVkVV2R5HTgCuAO4MiBobKjgA8DWwJnV9U5fflJwEeSrAV+DKweZX8kSZLGbeTTpkuF06aS\npKXIaVPNZtzTppIkSVoghjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJ\naojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSp\nIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSG\nGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpi\neJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojh\nTZIkqSGGN0mSpIYY3iRJkhoy0vCW5KQk65JcNlC2XZJzk3wnyeeSbDuw7tgka5NcmWT/gfJ9k1yW\n5KokJwyUb57ktH6bLyXZdZT9kSRJGrdRj7ydDBwwo+wY4LyqegxwPnAsQJK9gUOAvYADgROTpN/m\nvcARVbUnsGeS6X0eAdxcVY8GTgDeNsrOSJIkjdtIw1tVXQjcMqP4hcAp/fIpwMH98kHAaVV1Z1Vd\nA6wFViWZALauqkv6eqcObDO4rzOA5y94JyRJkpaQcZzztkNVrQOoqpuAHfrynYHrB+rd2JftDNww\nUH5DX3avbarqLuDWJNuPrumSJEnjtWLcDQBqAfeV+VauWbPmnuXJyUkmJycX8NCSJEkbZmpqiqmp\nqaHqjiO8rUuyY1Wt66dEf9iX3wg8cqDeLn3ZXOWD23w/yWbANlV181wHHgxvkiRJS8XMQaU3velN\nc9ZdjGnTcO8RsbOAV/bLhwFnDpSv7q8g3R3YA7i4n1q9Lcmq/gKGQ2dsc1i//BK6CyAkSZKWrVQt\n5KzljJ0nHwcmgYcB64DjgU8Dn6AbMbsWOKSqbu3rH0t3BekdwNFVdW5f/mTgw8CWwNlVdXRfvgXw\nEeBJwI+B1f3FDrO1pUbZV0mSNkQ3LtHa76fg79TRSkJVzXo62EjD21JieJMkLUWGN81mvvDmExYk\nSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMk\nSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJashQ4S3J\nbkl+s1/eKsnWo22WJEmSZnO/4S3Jq4EzgL/ri3YBPj3KRkmSJGl2w4y8HQU8C/h3gKpaC+wwykZJ\nkiRpdsOEt9ur6pfTb5KsAGp0TZIkSdJchglv/5TkOGCrJL8FfAL4zGibJUmSpNmkav5BtCQPAI4A\n9gcCfA74YN3fhktMktaaLEnaBCShvQmt4O/U0UpCVWXWdUOEtwcDv6iqu/r3mwFbVNXPF7ylI2R4\nkyQtRYY3zWa+8DbMtOkXgK0G3m8FnLcQDZMkSdL6GSa8bVlVP51+0y8/aHRNkiRJ0lyGCW8/S7Lv\n9JskTwb+Y3RNkiRJ0lxWDFHnj4BPJPk+3QULE8BLR9oqSZIkzep+L1gASPJA4DH92+9U1R0jbdUI\neMGCJGkp8oIFzWaDrjZN8ryqOj/Ji2ZbX1WfXMA2jpzhTZK0FBneNJv5wtt806b7AecDL5hlXQFN\nhTdJkqTlYN5p0/4Gvf+lqk5fvCaNhiNvkqSlyJE3zWaD7/NWVXcDfzqSVkmSJGm9DXOrkPOSvD7J\nI5NsP/0aecskSZJ0H8M8HuvqWYqrqh41miaNhtOmkqSlyGlTzWajnm26XBjeJElLkeFNs9nQq02n\nN94SOBJ4Nt1P1z8D76uqXyxoKyVJknS/hpk2PR34CfDRvuhlwEOr6iUjbtuCcuRNkrQUOfKm2WzU\nyBvw+Krae+D9BUmuWJimSZIkaX0Mc7XppUmePv0mydOAr4yuSZIkSZrLMNOmV9I91/S6vmhX4DvA\nnXRXne4z0hYuEKdNJUlLkdOmms3GTpv+9gK3R5IkSRvIW4VIkjRGjrxpNhv8eCxJkiQtLYY3SZKk\nhgwV3pLsluQ3++Wtkmy9sQdO8sdJvpnksiQfS7J5ku2SnJvkO0k+l2TbgfrHJlmb5Mok+w+U79vv\n46okJ2xsuyRJkpay+w1vSV4NnAH8XV+0C/DpjTlokp2A1wL79lerrgB+HzgGOK+qHgOcDxzb198b\nOATYCzgQODHdSQIA7wWOqKo9gT2THLAxbZMkSVrKhhl5Owp4FvDvAFW1FthhAY69GfDgJCuArYAb\ngRcCp/TrTwEO7pcPAk6rqjur6hpgLbAqyQSwdVVd0tc7dWAbSZKkZWeY8HZ7Vf1y+k0ftjbqEpOq\n+j7wN3T3jrsRuK2qzgN2rKp1fZ2b+FVI3Bm4fmAXN/ZlOwM3DJTf0JdJkiQtS8OEt39KchywVZLf\nAj4BfGZjDprkoXSjbLsBO9GNwP0B9w2FXocsSZI0YJib9B4DHAFcDvw/wNnABzfyuL8JfK+qbgZI\n8ingmcC6JDtW1bp+SvSHff0bgUcObL9LXzZX+azWrFlzz/Lk5CSTk5Mb2Q1JkqSNNzU1xdTU1FB1\nx3KT3iSrgJOApwK3AycDl9A9euvmqnprkj8DtquqY/oLFj4GPI1uWvTzwKOrqpJcBLyu3/4fgHdX\n1TmzHNOb9EqSlhxv0qvZbNTjsZJczn1/qm6jezj9/1dVP17fBlXVxUnOAL4G3NH/+X5ga+D0JIcD\n19JdYUpVXZHkdOCKvv6RA0nsKODDwJbA2bMFN0mSpOVimAfTvw24C/h4X7QaeBBwE/DsqnrBSFu4\nQBx5kyQtRY68aTbzjbwNE94urap9ZytLcnlVPWEB2zoyhjdJ0lJkeNNsNvbZppv156hN7+ypdPdo\nA7hzAdonSZKkIQ1ztekfAh9K8hAgdDfr/cMkDwb+apSNkyRJ0r0NfbXp9HNGq+q2kbZoRJw2lSQt\nRU6bajYbdbVpv4PfBR4HbDn9SNGqevOCtVCSJElDGebB9O8DXkr3IPkAL6F7MoIkSZIW2TBXm15W\nVfsM/PkQ4B+r6jmL08SF4bSpJGkpctpUs9nYq01/0f/58yQ70d0k9xEL1ThJkiQNb5hz3j7TP0j+\n7cCldP89+MBIWyVJkqRZzTttmuQBwNOr6ov9+y2ALVu84tRpU0nSUuS0qWazwdOmVXU38LcD729v\nMbhJkqTxmJhYSZKmXhMTK8f9sc1rmAsW3gF8Cfhky0NXjrxJkpai5T7yttz7N7IWbOSzTX8CPJju\n4fT/QXe7kKqqbRa6oaNkeJMkLUXLPdws9/6NrAUbc5Peqtp64ZskSZKkDTHMTXqT5OVJ3ti/f+Tg\ng+olSZK0eIa5z9uJwDOAl/Xvf8rARQySJElaPMPc5+1pVbVvkq8BVNUtSTYfcbskSZI0i2FG3u5I\nshn92YZJfg24e6StkiRJ0qyGCW/vBj4F7JDkL4ELgbeMtFWSJEma1f3eKgQgyWOB59PdJuQLVXXl\nqBu20LxViCRpKVrut9JY7v0bWQs28j5v7wZOm35EVqsMb5KkpWi5h5vl3r+RtWBDH4/V+yrw/yb5\n1yTvSPKUhW2eJEmShjXUtClAku2BFwOrgV2r6tGjbNhCc+RNkrQULfeRqeXev5G1YCNH3qbtATwW\n2A349kI0TJIkSetnmHPe3gb8Z+Bfgf8FfKqqbl2Eti0oR94kSUvRch+ZWu79G1kLNubZpnSh7RlV\n9aOFbZYkSZLW17C3CtkOeDSw5XRZVf2fEbZrwTnyJklaipb7yNRy79/IWrAxI29J/hA4GtgF+Drw\ndOBLwPMWspGSJEm6f8NcsHA08FTg2qp6LvAkoLlz3iRJkpaDYcLbL6rqFwBJtqiqbwOPGW2zJEmS\nNJthLli4IclDgU8Dn09yC3DtaJslSZKk2Qx9k16AJPsB2wLnVNUvR9aqEfCCBUnSUrTcT+hf7v0b\nWQs25tmmy4XhTZK0FC33cLPc+zeyFizQExYkSZI0ZoY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGG\nN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkrSkTUysJElTr4mJ\nleP+2LSMparG3YZFkaQ2lb5K0nKSBGjt3+8w7O8c+7cUDd+/kbUgoaoy27qxjbwl2TbJJ5JcmeRb\nSZ6WZLsk5yb5TpLPJdl2oP6xSdb29fcfKN83yWVJrkpywnh6I0mStDjGOW36LuDsqtoLeCLwbeAY\n4LyqegxwPnAsQJK9gUOAvYADgRPTRXmA9wJHVNWewJ5JDljcbkiSJC2esYS3JNsAz6mqkwGq6s6q\nug14IXBKX+0U4OB++SDgtL7eNcBaYFWSCWDrqrqkr3fqwDaSJEnLzrhG3nYHfpTk5CSXJnl/kgcB\nO1bVOoCqugnYoa+/M3D9wPY39mU7AzcMlN/Ql0mSJC1LK8Z43H2Bo6rqK0neSTdlOvPswAU9W3DN\nmjX3LE9OTjI5ObmQu5ckSdogU1NTTE1NDVV3LFebJtkR+FJVPap//2y68PbrwGRVreunRC+oqr2S\nHANUVb21r38OcDxw7XSdvnw1sF9VvWaWY3q1qSQ1aLlfrWj/liKvNr2Pfmr0+iR79kXPB74FnAW8\nsi87DDizXz4LWJ1k8yS7A3sAF/dTq7clWdVfwHDowDaSJEnLzrimTQFeB3wsyQOB7wGvAjYDTk9y\nON2o2iEAVXVFktOBK4A7gCMHhtGOAj4MbEl39eo5i9oLSZKkReRNeiVJS9pyn3azf0uR06aSJEla\nIIY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSG\nGN4kSZIb2ybmAAAOzUlEQVQaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGG\nN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTpGVgYmIlSZp5TUysHPdHJjUr\nVTXuNiyKJLWp9FXSpicJ0NK/cWHYf5Pb6xvYv4Gay7x/I2tBQlVltnWOvEmSJDXE8CZJktQQw5sk\nSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5Ik\nSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIk\nNcTwJkmS1JCxhrckD0hyaZKz+vfbJTk3yXeSfC7JtgN1j02yNsmVSfYfKN83yWVJrkpywjj6IUmS\ntFjGPfJ2NHDFwPtjgPOq6jHA+cCxAEn2Bg4B9gIOBE5Mkn6b9wJHVNWewJ5JDlisxkuSJC22sYW3\nJLsAvwN8cKD4hcAp/fIpwMH98kHAaVV1Z1VdA6wFViWZALauqkv6eqcObCNJ95iYWEmSZl4TEyvH\n/ZFJWqJWjPHY7wTeAGw7ULZjVa0DqKqbkuzQl+8MfGmg3o192Z3ADQPlN/TlknQv69ZdC9S4mzG0\ndety/5UkbZLGEt6S/C6wrqq+nmRynqoL+i/tmjVr7lmenJxkcnK+Q0uSJC2OqakppqamhqqbqsX/\nn2iStwAvpxs52wrYGvgU8BRgsqrW9VOiF1TVXkmOAaqq3tpvfw5wPHDtdJ2+fDWwX1W9ZpZj1jj6\nKmlp6E6TbenfgLA+/2Yt5/611zewfwM1l3n/RtaChKqadQh+LOe8VdVxVbVrVT0KWA2cX1WvAD4D\nvLKvdhhwZr98FrA6yeZJdgf2AC6uqpuA25Ks6i9gOHRgG0mSpGVnnOe8zeavgdOTHE43qnYIQFVd\nkeR0uitT7wCOHBhGOwr4MLAlcHZVnbPorZYkSVokY5k2HQenTaVNW3tTN06b3lOzub6B/Ruoucz7\nN7IWLLVpU0mSJG0Yw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTw\nJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmCYCJiZUk\naeo1MbFy3B+bJC26VNW427AoktSm0ldpQyQBWvs7Eob9e91e/4bvGyzv/rXXN7B/AzWXef9G1oKE\nqsps6xx5kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniT\nJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2S\nJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjepCFNTKwkSVOviYmV4/7YJEkLLFU1\n7jYsiiS1qfRVo5EEaO1nKAz7c2//lprh+wbLu3/t9Q3s30DNZd6/kbUgoaoy2zpH3iRJkhpieJMk\nSWqI4U2SJKkhhjdJkqSGjCW8JdklyflJvpXk8iSv68u3S3Juku8k+VySbQe2OTbJ2iRXJtl/oHzf\nJJcluSrJCePojyRJ0mIZ18jbncD/qKrHAc8AjkryWOAY4LyqegxwPnAsQJK9gUOAvYADgRPTXb4C\n8F7giKraE9gzyQGL2xVJkqTFM5bwVlU3VdXX++WfAlcCuwAvBE7pq50CHNwvHwScVlV3VtU1wFpg\nVZIJYOuquqSvd+rANpIkScvO2M95S7IS+A3gImDHqloHXcADduir7QxcP7DZjX3ZzsANA+U39GWS\nJEnL0opxHjzJQ4AzgKOr6qdJZt4Rb0HvkLdmzZp7licnJ5mcnFzI3UuSJG2Qqakppqamhqo7tics\nJFkBfBb4x6p6V192JTBZVev6KdELqmqvJMcAVVVv7eudAxwPXDtdpy9fDexXVa+Z5Xg+YUEbZbnf\nJdz+LTU+YeGems31DezfQM1l3r+RtWCJPmHhQ8AV08Gtdxbwyn75MODMgfLVSTZPsjuwB3BxP7V6\nW5JV/QUMhw5sI0mStOyMZeQtybOA/wNcThfHCzgOuBg4HXgk3ajaIVV1a7/NscARwB1006zn9uVP\nBj4MbAmcXVVHz3FMR960UZb7/x7t31LjyNs9NZvrG9i/gZrLvH8ja8E8I28+mF4a0nL/B8j+LTWG\nt3tqNtc3sH8DNZd5/0bWgiU6bSpJkqT1ZHiTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ404KZmFhJ\nkqZeExMrx/2xSZK0XrxViBbMcr8c3P4tRcu5f94q5J6azfUN7N9AzWXev5G1wFuFSJIkLQ+GN0mS\npIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeFllr\nz//02Z+SJC0tPtt08dtBW8948/l199S0f0vQcu6fzza9p2ZzfQP7N1BzmfdvZC3w2aaSJEnLg+FN\nkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhpieJMkSWqI4U2SJKkhhjdJ\nkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJ\nkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGrIswluS\n307y7SRXJfmzcbdnPKbG3QBtsKlxN0AbZWrcDdBGmRp3A7TBpsbdgLFpPrwleQDwHuAA4HHA7yd5\n7HhbNQ5T426ANtjUuBugjTI17gZoo0yNuwHaYFPjbsDYNB/egFXA2qq6tqruAE4DXjjmNkmSJI3E\ncghvOwPXD7y/oS+TJEladlJV427DRknyYuCAqvqv/fuXA6uq6nUz6rXdUUmStEmpqsxWvmKxGzIC\nNwK7DrzfpS+7l7k+AEmSpJYsh2nTS4A9kuyWZHNgNXDWmNskSZI0Es2PvFXVXUn+O3AuXRg9qaqu\nHHOzJEmSRqL5c94kSZI2Jcth2nST5g2K25VklyTnJ/lWksuTvO7+t9JSkuQBSS5N4qkajUmybZJP\nJLmy/zv4tHG3ScNL8sdJvpnksiQf60+b2mQY3hrmDYqbdyfwP6rqccAzgKP8/ppzNHDFuBuhDfIu\n4Oyq2gt4IuDpNo1IshPwWmDfqtqH7hSw1eNt1eIyvLXNGxQ3rKpuqqqv98s/pfvl4T0KG5FkF+B3\ngA+Ouy1aP0m2AZ5TVScDVNWdVfXvY26W1s9mwIOTrAAeBHx/zO1ZVIa3tnmD4mUiyUrgN4Avj7cl\nWg/vBN4AeOJwe3YHfpTk5H7a+/1Jthp3ozScqvo+8DfAdXS3Bru1qs4bb6sWl+FNGrMkDwHOAI7u\nR+C0xCX5XWBdP3Ka/qV2rAD2Bf62qvYFfg4cM94maVhJHko3y7QbsBPwkCQvG2+rFpfhrW1D3aBY\nS1c/5H8G8JGqOnPc7dHQngUclOR7wN8Dz01y6pjbpOHdAFxfVV/p359BF+bUht8EvldVN1fVXcAn\ngWeOuU2LyvDWNm9Q3L4PAVdU1bvG3RANr6qOq6pdq+pRdH/vzq+qQ8fdLg2nqtYB1yfZsy96Pl54\n0pLrgKcn2TJJ6L6/TeqCk+Zv0rsp8wbFbUvyLOAPgMuTfI3u3Knjquqc8bZM2iS8DvhYkgcC3wNe\nNeb2aEhVdXGSM4CvAXf0f75/vK1aXN6kV5IkqSFOm0qSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJ\nkiQ1xPAmSZLUEMObpGUvyQVJFuQO+v1NsS8f1/aSZHiTpHkk2WyW4o29QeZ9tp/jOJJ0H4Y3SUtC\nPyJ1RZL3J/lmknOSbNGvu2fkLMnDklzdLx+W5FNJzk3yvSRHJfnjJJcm+WL/AOtphyb5WpLLkjy1\n3/5BSU5KclGSryZ5wcB+z0zyBeC8WZr7wCQf7dt7epIt++32TTKV5JIk/5hkx778yUm+3j9J46iB\nPt/nOEnenuTyJN9IcshA3fuUJ9mvP96nk3w3yV8leVmSL/f1du/rvaTf9mtJphbi+5I0PoY3SUvJ\nHsD/rKrHA7cBL56j3uDI1eOAg4FVwF8CP62qfYGLgMHnjW5VVU+iC08f6sv+HPhCVT0deB7wjiRb\n9eueBLyoqp47y/EfA7ynqvYGfgIcmWQF8D+BF1fVU4GTgbf09T8EHNUff6Z7jpPkRcA+VfUE4LeA\ntyfZca7yfvt9gP8K7A28Anh0VT0NOAl4bV/njcD+/fEPmqUNkhpieJO0lFxdVdPng30VWDnENhdU\n1c+r6kfArcBn+/LLZ2z/9wBV9c/A1km2AfYHjulHxKaAzYFd+/qfr6rb5jjmdVV1Ub/8UeDZdIHu\n8cDn+/39ObBTkm2BbavqX/r6H5mxr8HjPHugnT/s27RqjvKn9ttcUlU/rKpfAv9K96zjmf2/EDgl\nyR/iM62l5vmXWNJScvvA8l3Alv3ynfzqP5tbcm+D29TA+7u5979xM88zKyB0I2VrB1ckeTrws3na\nOde+vllVz5qxr23n2Q/3c5zMcqzp8mmD/b+bWfpfVUf2U8W/B3w1yb5Vdcv9tEvSEuXIm6SlJHOU\nXwM8pV9+yQbu+6UASZ4N3FZVPwE+B7zunoMnvzHkvnZL8rR++WXAPwPfAX6tD34kWZFk735U7dYk\nz+zrv3ye/f4z8NIkD0jya8BzgIvnKR9KkkdV1SVVdTzwQ+CRw24raekxvElaSua6ivMdwGuSfBXY\nfgO2L+AXSS4FTgQO78v/gu7ig8uSfBN485Dt/DZwVJIrgIcC76uqO4D/Arw1ydeBrwHP6OsfDpzY\nH3/OK1Wr6lPAZcA36C5geEM/JTpr+Rz9nM3b+z5eBvxLVV02ZD8lLUGp2tgr3iVJkrRYHHmTJElq\niOFNkiSpIYY3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIb8X6UhIC+FANsJAAAAAElFTkSu\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11faf6850>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plots to visualise results by number of bedrooms.\n",
    "plt.figure(num=1,figsize=(10,8))\n",
    "plt.bar(df_by_bedroom['number_bedrooms'],df_by_bedroom['count'],align='center')\n",
    "plt.xlabel('number bedrooms')\n",
    "plt.ylabel('count')\n",
    "plt.title('count by number of bedrooms')\n",
    "plt.ticklabel_format(style='plain')\n",
    "\n",
    "plt.figure(num=2,figsize=(10,8))\n",
    "plt.bar(df_by_bedroom['number_bedrooms'],df_by_bedroom['average_price'],align='center')\n",
    "plt.xlabel('number bedrooms')\n",
    "plt.ylabel('average price')\n",
    "plt.title('average price by number of bedrooms')\n",
    "plt.ticklabel_format(style='plain')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using the postcode csv to find the property borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Read in the csv\n",
    "london_postcodes_df = pd.read_csv('london_postcodes.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Postcode</th>\n",
       "      <th>District</th>\n",
       "      <th>Built_up_sub_division</th>\n",
       "      <th>London_Borough</th>\n",
       "      <th>Inner_Outer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BR1 1AA</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BR1 1AB</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BR1 1AE</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BR1 1AX</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BR1 1BA</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Postcode             District Built_up_sub_division London_Borough  \\\n",
       "0  BR1 1AA  Bromley London Boro               Bromley        Bromley   \n",
       "1  BR1 1AB  Bromley London Boro               Bromley        Bromley   \n",
       "2  BR1 1AE  Bromley London Boro               Bromley        Bromley   \n",
       "3  BR1 1AX  Bromley London Boro               Bromley        Bromley   \n",
       "4  BR1 1BA  Bromley London Boro               Bromley        Bromley   \n",
       "\n",
       "  Inner_Outer  \n",
       "0       Outer  \n",
       "1       Outer  \n",
       "2       Outer  \n",
       "3       Outer  \n",
       "4       Outer  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Tidy up column names\n",
    "def Replace_Bad_Characters_in_Cols(df):\n",
    "    to_replace = (\" \",\")\",\"(\",\":\",\"/\",\"%\",\"-\")\n",
    "    for i in to_replace:\n",
    "        df.columns = df.columns.str.replace(i, \"_\")\n",
    "Replace_Bad_Characters_in_Cols(london_postcodes_df)\n",
    "london_postcodes_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Postcode</th>\n",
       "      <th>District</th>\n",
       "      <th>Built_up_sub_division</th>\n",
       "      <th>London_Borough</th>\n",
       "      <th>Inner_Outer</th>\n",
       "      <th>pc_stem</th>\n",
       "      <th>pc_end</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BR1 1AA</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "      <td>BR1</td>\n",
       "      <td>1AA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BR1 1AB</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "      <td>BR1</td>\n",
       "      <td>1AB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BR1 1AE</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "      <td>BR1</td>\n",
       "      <td>1AE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BR1 1AX</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "      <td>BR1</td>\n",
       "      <td>1AX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BR1 1BA</td>\n",
       "      <td>Bromley London Boro</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Bromley</td>\n",
       "      <td>Outer</td>\n",
       "      <td>BR1</td>\n",
       "      <td>1BA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Postcode             District Built_up_sub_division London_Borough  \\\n",
       "0  BR1 1AA  Bromley London Boro               Bromley        Bromley   \n",
       "1  BR1 1AB  Bromley London Boro               Bromley        Bromley   \n",
       "2  BR1 1AE  Bromley London Boro               Bromley        Bromley   \n",
       "3  BR1 1AX  Bromley London Boro               Bromley        Bromley   \n",
       "4  BR1 1BA  Bromley London Boro               Bromley        Bromley   \n",
       "\n",
       "  Inner_Outer pc_stem pc_end  \n",
       "0       Outer     BR1    1AA  \n",
       "1       Outer     BR1    1AB  \n",
       "2       Outer     BR1    1AE  \n",
       "3       Outer     BR1    1AX  \n",
       "4       Outer     BR1    1BA  "
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create new columns for the stem & end of the postcodes.\n",
    "postcode_split = london_postcodes_df.Postcode.str.split(' ', expand=True).astype(str)\n",
    "postcode_split.columns=('pc_stem','pc_end')\n",
    "pc_df = pd.merge(london_postcodes_df,postcode_split,how='left',left_index=True,right_index=True)\n",
    "pc_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>London_Borough</th>\n",
       "      <th>Barking and Dagenham</th>\n",
       "      <th>Barnet</th>\n",
       "      <th>Bexley</th>\n",
       "      <th>Brent</th>\n",
       "      <th>Bromley</th>\n",
       "      <th>Camden</th>\n",
       "      <th>City of London</th>\n",
       "      <th>Croydon</th>\n",
       "      <th>Ealing</th>\n",
       "      <th>Enfield</th>\n",
       "      <th>...</th>\n",
       "      <th>Merton</th>\n",
       "      <th>Newham</th>\n",
       "      <th>Redbridge</th>\n",
       "      <th>Richmond upon Thames</th>\n",
       "      <th>Southwark</th>\n",
       "      <th>Sutton</th>\n",
       "      <th>Tower Hamlets</th>\n",
       "      <th>Waltham Forest</th>\n",
       "      <th>Wandsworth</th>\n",
       "      <th>Westminster</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pc_stem</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BR1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1724.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BR2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1383.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BR3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2027.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BR4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>557.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BR5</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1292.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "London_Borough  Barking and Dagenham  Barnet  Bexley  Brent  Bromley  Camden  \\\n",
       "pc_stem                                                                        \n",
       "BR1                              NaN     NaN     NaN    NaN   1724.0     NaN   \n",
       "BR2                              NaN     NaN     NaN    NaN   1383.0     NaN   \n",
       "BR3                              NaN     NaN     NaN    NaN   2027.0     NaN   \n",
       "BR4                              NaN     NaN     NaN    NaN    557.0     NaN   \n",
       "BR5                              NaN     NaN     NaN    NaN   1292.0     NaN   \n",
       "\n",
       "London_Borough  City of London  Croydon  Ealing  Enfield     ...       Merton  \\\n",
       "pc_stem                                                      ...                \n",
       "BR1                        NaN      NaN     NaN      NaN     ...          NaN   \n",
       "BR2                        NaN      NaN     NaN      NaN     ...          NaN   \n",
       "BR3                        NaN      1.0     NaN      NaN     ...          NaN   \n",
       "BR4                        NaN      NaN     NaN      NaN     ...          NaN   \n",
       "BR5                        NaN      NaN     NaN      NaN     ...          NaN   \n",
       "\n",
       "London_Borough  Newham  Redbridge  Richmond upon Thames  Southwark  Sutton  \\\n",
       "pc_stem                                                                      \n",
       "BR1                NaN        NaN                   NaN        NaN     NaN   \n",
       "BR2                NaN        NaN                   NaN        NaN     NaN   \n",
       "BR3                NaN        NaN                   NaN        NaN     NaN   \n",
       "BR4                NaN        NaN                   NaN        NaN     NaN   \n",
       "BR5                NaN        NaN                   NaN        NaN     NaN   \n",
       "\n",
       "London_Borough  Tower Hamlets  Waltham Forest  Wandsworth  Westminster  \n",
       "pc_stem                                                                 \n",
       "BR1                       NaN             NaN         NaN          NaN  \n",
       "BR2                       NaN             NaN         NaN          NaN  \n",
       "BR3                       NaN             NaN         NaN          NaN  \n",
       "BR4                       NaN             NaN         NaN          NaN  \n",
       "BR5                       NaN             NaN         NaN          NaN  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a pivot table of the postcode stem by how many full postcodes sit in each borough\n",
    "pivot = pd.DataFrame(pd.pivot_table(pc_df,values='Postcode',index = 'pc_stem',columns='London_Borough',aggfunc='count'))\n",
    "pivot.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately many postcode stems have full postcodes in more than one borough; e.g. N1 has postcodes in both Islington and Hackney. We'll just default to whichever borough has the greatest number of postcodes for each postcode stem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Create dataframe of postcode stems & boroughs\n",
    "pc_borough_lookup = pd.DataFrame(pivot.idxmax(axis=1))\n",
    "pc_borough_lookup.reset_index( inplace=True)\n",
    "\n",
    "# Merge borough column into main df\n",
    "df = pd.merge(df,pc_borough_lookup,how='left',left_on='postcode',right_on='pc_stem')\n",
    "df = df.rename(columns={0:'London_Borough'})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "i_o_df = pd.DataFrame(pd.pivot_table(pc_df,values='Postcode',index ='London_Borough',columns='Inner_Outer',aggfunc='count'))\n",
    "i_o_df.Inner = i_o_df.Inner/i_o_df.Inner # change to 1s\n",
    "i_o_df.Outer = i_o_df.Outer/i_o_df.Outer # change to 1s\n",
    "i_o_df.rename(columns={'Inner_Outer':'London_Borough'},inplace=True)\n",
    "i_o_lookup = pd.DataFrame(i_o_df.idxmax(axis=1)) # Create lookup table of borough - inner/outer\n",
    "i_o_lookup.rename(columns={0:'Inner_Outer'},inplace=True)\n",
    "i_o_lookup.reset_index(inplace=True)\n",
    "df = pd.merge(df,i_o_lookup,how='left',on='London_Borough')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>type</th>\n",
       "      <th>address</th>\n",
       "      <th>url</th>\n",
       "      <th>postcode</th>\n",
       "      <th>number_bedrooms</th>\n",
       "      <th>search_date</th>\n",
       "      <th>pc_stem</th>\n",
       "      <th>London_Borough</th>\n",
       "      <th>Inner_Outer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3012.0</td>\n",
       "      <td>2 bedroom flat</td>\n",
       "      <td>Horsley Court, Montaigne Close, London</td>\n",
       "      <td>/property-to-rent/property-44782458.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1100.0</td>\n",
       "      <td>1 bedroom flat</td>\n",
       "      <td>Beaconsfield Road, London</td>\n",
       "      <td>/property-to-rent/property-44818353.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1450.0</td>\n",
       "      <td>2 bedroom flat</td>\n",
       "      <td>Gaumont House,  Marmont Road,  London</td>\n",
       "      <td>/property-to-rent/property-44818389.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1257.0</td>\n",
       "      <td>3 bedroom flat</td>\n",
       "      <td>Lady Hamilton Court,  Holders Hill Avenue,  Lo...</td>\n",
       "      <td>/property-to-rent/property-44818368.html</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1300.0</td>\n",
       "      <td>1 bedroom flat</td>\n",
       "      <td>Elm Park, London, SW2</td>\n",
       "      <td>/property-to-rent/property-56484454.html</td>\n",
       "      <td>SW2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>11/10/2016</td>\n",
       "      <td>SW2</td>\n",
       "      <td>Lambeth</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    price            type                                            address  \\\n",
       "0  3012.0  2 bedroom flat             Horsley Court, Montaigne Close, London   \n",
       "1  1100.0  1 bedroom flat                          Beaconsfield Road, London   \n",
       "2  1450.0  2 bedroom flat              Gaumont House,  Marmont Road,  London   \n",
       "3  1257.0  3 bedroom flat  Lady Hamilton Court,  Holders Hill Avenue,  Lo...   \n",
       "4  1300.0  1 bedroom flat                              Elm Park, London, SW2   \n",
       "\n",
       "                                        url postcode  number_bedrooms  \\\n",
       "0  /property-to-rent/property-44782458.html      NaN              2.0   \n",
       "1  /property-to-rent/property-44818353.html      NaN              1.0   \n",
       "2  /property-to-rent/property-44818389.html      NaN              2.0   \n",
       "3  /property-to-rent/property-44818368.html      NaN              3.0   \n",
       "4  /property-to-rent/property-56484454.html      SW2              1.0   \n",
       "\n",
       "  search_date pc_stem London_Borough Inner_Outer  \n",
       "0  11/10/2016     NaN            NaN         NaN  \n",
       "1  11/10/2016     NaN            NaN         NaN  \n",
       "2  11/10/2016     NaN            NaN         NaN  \n",
       "3  11/10/2016     NaN            NaN         NaN  \n",
       "4  11/10/2016     SW2        Lambeth       Inner  "
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of records in full dataframe:  4971\n",
      "number of records with borough data:  3639\n"
     ]
    }
   ],
   "source": [
    "print 'number of records in full dataframe: ',len(df)\n",
    "print 'number of records with borough data: ',len(df[df.London_Borough.notnull()])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Focus on 1 beds by borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>London_Borough</th>\n",
       "      <th>number of property listings</th>\n",
       "      <th>mean price</th>\n",
       "      <th>Inner_Outer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Kensington and Chelsea</td>\n",
       "      <td>85</td>\n",
       "      <td>2345.611765</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>152</td>\n",
       "      <td>2249.532895</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>City of London</td>\n",
       "      <td>4</td>\n",
       "      <td>2118.250000</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Camden</td>\n",
       "      <td>68</td>\n",
       "      <td>1755.235294</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Islington</td>\n",
       "      <td>79</td>\n",
       "      <td>1700.759494</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Southwark</td>\n",
       "      <td>59</td>\n",
       "      <td>1677.881356</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Hammersmith and Fulham</td>\n",
       "      <td>36</td>\n",
       "      <td>1644.666667</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Hackney</td>\n",
       "      <td>18</td>\n",
       "      <td>1576.277778</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Tower Hamlets</td>\n",
       "      <td>135</td>\n",
       "      <td>1570.696296</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Hounslow</td>\n",
       "      <td>7</td>\n",
       "      <td>1502.428571</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Brent</td>\n",
       "      <td>26</td>\n",
       "      <td>1496.384615</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Greenwich</td>\n",
       "      <td>28</td>\n",
       "      <td>1487.500000</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Wandsworth</td>\n",
       "      <td>42</td>\n",
       "      <td>1484.309524</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Lambeth</td>\n",
       "      <td>48</td>\n",
       "      <td>1434.125000</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Merton</td>\n",
       "      <td>14</td>\n",
       "      <td>1378.000000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Ealing</td>\n",
       "      <td>26</td>\n",
       "      <td>1372.000000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Richmond upon Thames</td>\n",
       "      <td>3</td>\n",
       "      <td>1366.666667</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Haringey</td>\n",
       "      <td>28</td>\n",
       "      <td>1358.750000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Newham</td>\n",
       "      <td>45</td>\n",
       "      <td>1344.288889</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Barnet</td>\n",
       "      <td>38</td>\n",
       "      <td>1273.578947</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Lewisham</td>\n",
       "      <td>27</td>\n",
       "      <td>1132.666667</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Waltham Forest</td>\n",
       "      <td>16</td>\n",
       "      <td>1078.000000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Croydon</td>\n",
       "      <td>14</td>\n",
       "      <td>1041.071429</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Enfield</td>\n",
       "      <td>15</td>\n",
       "      <td>1021.000000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Bromley</td>\n",
       "      <td>5</td>\n",
       "      <td>965.000000</td>\n",
       "      <td>Outer</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            London_Borough  number of property listings   mean price  \\\n",
       "0   Kensington and Chelsea                           85  2345.611765   \n",
       "1              Westminster                          152  2249.532895   \n",
       "2           City of London                            4  2118.250000   \n",
       "3                   Camden                           68  1755.235294   \n",
       "4                Islington                           79  1700.759494   \n",
       "5                Southwark                           59  1677.881356   \n",
       "6   Hammersmith and Fulham                           36  1644.666667   \n",
       "7                  Hackney                           18  1576.277778   \n",
       "8            Tower Hamlets                          135  1570.696296   \n",
       "9                 Hounslow                            7  1502.428571   \n",
       "10                   Brent                           26  1496.384615   \n",
       "11               Greenwich                           28  1487.500000   \n",
       "12              Wandsworth                           42  1484.309524   \n",
       "13                 Lambeth                           48  1434.125000   \n",
       "14                  Merton                           14  1378.000000   \n",
       "15                  Ealing                           26  1372.000000   \n",
       "16    Richmond upon Thames                            3  1366.666667   \n",
       "17                Haringey                           28  1358.750000   \n",
       "18                  Newham                           45  1344.288889   \n",
       "19                  Barnet                           38  1273.578947   \n",
       "20                Lewisham                           27  1132.666667   \n",
       "21          Waltham Forest                           16  1078.000000   \n",
       "22                 Croydon                           14  1041.071429   \n",
       "23                 Enfield                           15  1021.000000   \n",
       "24                 Bromley                            5   965.000000   \n",
       "\n",
       "   Inner_Outer  \n",
       "0        Inner  \n",
       "1        Inner  \n",
       "2        Inner  \n",
       "3        Inner  \n",
       "4        Inner  \n",
       "5        Inner  \n",
       "6        Inner  \n",
       "7        Inner  \n",
       "8        Inner  \n",
       "9        Outer  \n",
       "10       Outer  \n",
       "11       Inner  \n",
       "12       Inner  \n",
       "13       Inner  \n",
       "14       Outer  \n",
       "15       Outer  \n",
       "16       Outer  \n",
       "17       Outer  \n",
       "18       Outer  \n",
       "19       Outer  \n",
       "20       Inner  \n",
       "21       Outer  \n",
       "22       Outer  \n",
       "23       Outer  \n",
       "24       Outer  "
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1beds = df[ (df['London_Borough'].notnull()) & (df['number_bedrooms']==1.0)]\n",
    "borough_pivot = pd.pivot_table(data=df_1beds,values='price',index='London_Borough',aggfunc=('count','mean'))\n",
    "borough_pivot.sort_values(by='mean',inplace=True,ascending=False)\n",
    "borough_pivot.rename(columns={'count':'number of property listings','mean':'mean price'},inplace=True)\n",
    "borough_pivot.reset_index(inplace=True)\n",
    "one_beds = pd.merge(borough_pivot,i_o_lookup,how='left',on='London_Borough')\n",
    "one_beds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now easily see the cheapest inner London borough to rent!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>London_Borough</th>\n",
       "      <th>number of property listings</th>\n",
       "      <th>mean price</th>\n",
       "      <th>Inner_Outer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Lewisham</td>\n",
       "      <td>27</td>\n",
       "      <td>1132.666667</td>\n",
       "      <td>Inner</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   London_Borough  number of property listings   mean price Inner_Outer\n",
       "20       Lewisham                           27  1132.666667       Inner"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "last_inner_row=len(one_beds[(one_beds['Inner_Outer']=='Inner')])\n",
    "one_beds[(one_beds['Inner_Outer']=='Inner')][last_inner_row-1:]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}