ML - Datasets and Questions - Enron Dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# What is a person of interest?\n",
    "* Indicted\n",
    "* Settled without admitting guilt\n",
    "* Testified in exchange for Immunity\n",
    "\n",
    "### Accuracy <=> Training Set Size\n",
    "* MORE DATA > FINE TUNED ALGORITHM!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\"\"\" \n",
    "    Starter code for exploring the Enron dataset (emails + finances);\n",
    "    loads up the dataset (pickled dict of dicts).\n",
    "\n",
    "    The dataset has the form:\n",
    "    enron_data[\"LASTNAME FIRSTNAME MIDDLEINITIAL\"] = { features_dict }\n",
    "\n",
    "    {features_dict} is a dictionary of features associated with that person.\n",
    "    You should explore features_dict as part of the mini-project,\n",
    "    but here's an example to get you started:\n",
    "\n",
    "    enron_data[\"SKILLING JEFFREY K\"][\"bonus\"] = 5600000\n",
    "    \n",
    "\"\"\"\n",
    "\n",
    "import pickle\n",
    "\n",
    "enron_data = pickle.load(open(\"C:/Users/nmeli/Documents/Udacity/Data Analysis/Notebooks/Machine Learning/ud120-projects/final_project/final_project_dataset.pkl\", \"r\"))\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5600000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "enron_data[\"SKILLING JEFFREY K\"][\"bonus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bonus': 5600000,\n",
       " 'deferral_payments': 'NaN',\n",
       " 'deferred_income': 'NaN',\n",
       " 'director_fees': 'NaN',\n",
       " 'email_address': 'jeff.skilling@enron.com',\n",
       " 'exercised_stock_options': 19250000,\n",
       " 'expenses': 29336,\n",
       " 'from_messages': 108,\n",
       " 'from_poi_to_this_person': 88,\n",
       " 'from_this_person_to_poi': 30,\n",
       " 'loan_advances': 'NaN',\n",
       " 'long_term_incentive': 1920000,\n",
       " 'other': 22122,\n",
       " 'poi': True,\n",
       " 'restricted_stock': 6843672,\n",
       " 'restricted_stock_deferred': 'NaN',\n",
       " 'salary': 1111258,\n",
       " 'shared_receipt_with_poi': 2042,\n",
       " 'to_messages': 3627,\n",
       " 'total_payments': 8682716,\n",
       " 'total_stock_value': 26093672}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "enron_data[\"SKILLING JEFFREY K\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "146"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How many data points(people) are in the dataset?\n",
    "len(enron_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "METTS MARK\n",
      "BAXTER JOHN C\n",
      "ELLIOTT STEVEN\n",
      "CORDES WILLIAM R\n",
      "HANNON KEVIN P\n",
      "MORDAUNT KRISTINA M\n",
      "MEYER ROCKFORD G\n",
      "MCMAHON JEFFREY\n",
      "HORTON STANLEY C\n",
      "PIPER GREGORY F\n",
      "HUMPHREY GENE E\n",
      "UMANOFF ADAM S\n",
      "BLACHMAN JEREMY M\n",
      "SUNDE MARTIN\n",
      "GIBBS DANA R\n",
      "LOWRY CHARLES P\n",
      "COLWELL WESLEY\n",
      "MULLER MARK S\n",
      "JACKSON CHARLENE R\n",
      "WESTFAHL RICHARD K\n",
      "WALTERS GARETH W\n",
      "WALLS JR ROBERT H\n",
      "KITCHEN LOUISE\n",
      "CHAN RONNIE\n",
      "BELFER ROBERT\n",
      "SHANKMAN JEFFREY A\n",
      "WODRASKA JOHN\n",
      "BERGSIEKER RICHARD P\n",
      "URQUHART JOHN A\n",
      "BIBI PHILIPPE A\n",
      "RIEKER PAULA H\n",
      "WHALEY DAVID A\n",
      "BECK SALLY W\n",
      "HAUG DAVID L\n",
      "ECHOLS JOHN B\n",
      "MENDELSOHN JOHN\n",
      "HICKERSON GARY J\n",
      "CLINE KENNETH W\n",
      "LEWIS RICHARD\n",
      "HAYES ROBERT E\n",
      "MCCARTY DANNY J\n",
      "KOPPER MICHAEL J\n",
      "LEFF DANIEL P\n",
      "LAVORATO JOHN J\n",
      "BERBERIAN DAVID\n",
      "DETMERING TIMOTHY J\n",
      "WAKEHAM JOHN\n",
      "POWERS WILLIAM\n",
      "GOLD JOSEPH\n",
      "BANNANTINE JAMES M\n",
      "DUNCAN JOHN H\n",
      "SHAPIRO RICHARD S\n",
      "SHERRIFF JOHN R\n",
      "SHELBY REX\n",
      "LEMAISTRE CHARLES\n",
      "DEFFNER JOSEPH M\n",
      "KISHKILL JOSEPH G\n",
      "WHALLEY LAWRENCE G\n",
      "MCCONNELL MICHAEL S\n",
      "PIRO JIM\n",
      "DELAINEY DAVID W\n",
      "SULLIVAN-SHAKLOVITZ COLLEEN\n",
      "WROBEL BRUCE\n",
      "LINDHOLM TOD A\n",
      "MEYER JEROME J\n",
      "LAY KENNETH L\n",
      "BUTTS ROBERT H\n",
      "OLSON CINDY K\n",
      "MCDONALD REBECCA\n",
      "CUMBERLAND MICHAEL S\n",
      "GAHN ROBERT S\n",
      "MCCLELLAN GEORGE\n",
      "HERMANN ROBERT J\n",
      "SCRIMSHAW MATTHEW\n",
      "GATHMANN WILLIAM D\n",
      "HAEDICKE MARK E\n",
      "BOWEN JR RAYMOND M\n",
      "GILLIS JOHN\n",
      "FITZGERALD JAY L\n",
      "MORAN MICHAEL P\n",
      "REDMOND BRIAN L\n",
      "BAZELIDES PHILIP J\n",
      "BELDEN TIMOTHY N\n",
      "DURAN WILLIAM D\n",
      "THORN TERENCE H\n",
      "FASTOW ANDREW S\n",
      "FOY JOE\n",
      "CALGER CHRISTOPHER F\n",
      "RICE KENNETH D\n",
      "KAMINSKI WINCENTY J\n",
      "LOCKHART EUGENE E\n",
      "COX DAVID\n",
      "OVERDYKE JR JERE C\n",
      "PEREIRA PAULO V. FERRAZ\n",
      "STABLER FRANK\n",
      "SKILLING JEFFREY K\n",
      "BLAKE JR. NORMAN P\n",
      "SHERRICK JEFFREY B\n",
      "PRENTICE JAMES\n",
      "GRAY RODNEY\n",
      "PICKERING MARK R\n",
      "THE TRAVEL AGENCY IN THE PARK\n",
      "NOLES JAMES L\n",
      "KEAN STEVEN J\n",
      "TOTAL\n",
      "FOWLER PEGGY\n",
      "WASAFF GEORGE\n",
      "WHITE JR THOMAS E\n",
      "CHRISTODOULOU DIOMEDES\n",
      "ALLEN PHILLIP K\n",
      "SHARP VICTORIA T\n",
      "JAEDICKE ROBERT\n",
      "WINOKUR JR. HERBERT S\n",
      "BROWN MICHAEL\n",
      "BADUM JAMES P\n",
      "HUGHES JAMES A\n",
      "REYNOLDS LAWRENCE\n",
      "DIMICHELE RICHARD G\n",
      "BHATNAGAR SANJAY\n",
      "CARTER REBECCA C\n",
      "BUCHANAN HAROLD G\n",
      "YEAP SOON\n",
      "MURRAY JULIA H\n",
      "GARLAND C KEVIN\n",
      "DODSON KEITH\n",
      "YEAGER F SCOTT\n",
      "HIRKO JOSEPH\n",
      "DIETRICH JANET R\n",
      "DERRICK JR. JAMES V\n",
      "FREVERT MARK A\n",
      "PAI LOU L\n",
      "BAY FRANKLIN R\n",
      "HAYSLETT RODERICK J\n",
      "FUGH JOHN L\n",
      "FALLON JAMES B\n",
      "KOENIG MARK E\n",
      "SAVAGE FRANK\n",
      "IZZO LAWRENCE L\n",
      "TILNEY ELIZABETH A\n",
      "MARTIN AMANDA K\n",
      "BUY RICHARD B\n",
      "GRAMM WENDY L\n",
      "CAUSEY RICHARD A\n",
      "TAYLOR MITCHELL S\n",
      "DONAHUE JR JEFFREY M\n",
      "GLISAN JR BEN F\n"
     ]
    }
   ],
   "source": [
    "for n in enron_data:\n",
    "    print n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n",
      "21\n"
     ]
    }
   ],
   "source": [
    "# For each person, how many features are available?\n",
    "for k,v in enron_data.items():\n",
    "    print(len(v.keys()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18\n"
     ]
    }
   ],
   "source": [
    "# How many POIs are there in the E+F dataset\n",
    "count = 0\n",
    "for p in enron_data:\n",
    "    for i in enron_data[p]:\n",
    "        if i == \"poi\":\n",
    "            if enron_data[p][i] == 1:\n",
    "                count +=1\n",
    "print count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1095040"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What is the total value of the stock belonging to James Prentice?\n",
    "enron_data[\"PRENTICE JAMES\"]['total_stock_value']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bonus': 1200000,\n",
       " 'deferral_payments': 27610,\n",
       " 'deferred_income': -144062,\n",
       " 'director_fees': 'NaN',\n",
       " 'email_address': 'wes.colwell@enron.com',\n",
       " 'exercised_stock_options': 'NaN',\n",
       " 'expenses': 16514,\n",
       " 'from_messages': 40,\n",
       " 'from_poi_to_this_person': 240,\n",
       " 'from_this_person_to_poi': 11,\n",
       " 'loan_advances': 'NaN',\n",
       " 'long_term_incentive': 'NaN',\n",
       " 'other': 101740,\n",
       " 'poi': True,\n",
       " 'restricted_stock': 698242,\n",
       " 'restricted_stock_deferred': 'NaN',\n",
       " 'salary': 288542,\n",
       " 'shared_receipt_with_poi': 1132,\n",
       " 'to_messages': 1758,\n",
       " 'total_payments': 1490344,\n",
       " 'total_stock_value': 698242}"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How many email messages do we have from Wesley Colwell to persons of interest?\n",
    "enron_data[\"COLWELL WESLEY\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bonus': 5600000,\n",
       " 'deferral_payments': 'NaN',\n",
       " 'deferred_income': 'NaN',\n",
       " 'director_fees': 'NaN',\n",
       " 'email_address': 'jeff.skilling@enron.com',\n",
       " 'exercised_stock_options': 19250000,\n",
       " 'expenses': 29336,\n",
       " 'from_messages': 108,\n",
       " 'from_poi_to_this_person': 88,\n",
       " 'from_this_person_to_poi': 30,\n",
       " 'loan_advances': 'NaN',\n",
       " 'long_term_incentive': 1920000,\n",
       " 'other': 22122,\n",
       " 'poi': True,\n",
       " 'restricted_stock': 6843672,\n",
       " 'restricted_stock_deferred': 'NaN',\n",
       " 'salary': 1111258,\n",
       " 'shared_receipt_with_poi': 2042,\n",
       " 'to_messages': 3627,\n",
       " 'total_payments': 8682716,\n",
       " 'total_stock_value': 26093672}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What’s the value of stock options exercised by Jeffrey K Skilling?\n",
    "enron_data[\"SKILLING JEFFREY K\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8682716\n",
      "2424083\n",
      "103559793\n"
     ]
    }
   ],
   "source": [
    "# Of these three individuals (Lay, Skilling and Fastow), who took home the most money (largest value of “total_payments” feature)?\n",
    "print enron_data[\"SKILLING JEFFREY K\"]['total_payments']\n",
    "print enron_data[\"FASTOW ANDREW S\"]['total_payments']\n",
    "print enron_data[\"LAY KENNETH L\"]['total_payments']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95\n",
      "111\n"
     ]
    }
   ],
   "source": [
    "# How many folks in this dataset have a quantified salary? What about a known email address?\n",
    "count1 = 0\n",
    "count2 = 0 \n",
    "for p in enron_data:\n",
    "    for i in enron_data[p]:\n",
    "        if i == \"salary\":\n",
    "            temp1 = enron_data[p][i]\n",
    "            #print temp\n",
    "            if temp1 =='NaN':\n",
    "                pass\n",
    "            else:\n",
    "                count1 +=1\n",
    "        if i == \"email_address\":\n",
    "            temp2 = enron_data[p][i]\n",
    "            if temp2 =='NaN':\n",
    "                pass\n",
    "            else:\n",
    "                count2 +=1\n",
    "print count1\n",
    "print count2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\"\"\" \n",
    "    A general tool for converting data from the\n",
    "    dictionary format to an (n x k) python list that's \n",
    "    ready for training an sklearn algorithm\n",
    "\n",
    "    n--no. of key-value pairs in dictonary\n",
    "    k--no. of features being extracted\n",
    "\n",
    "    dictionary keys are names of persons in dataset\n",
    "    dictionary values are dictionaries, where each\n",
    "        key-value pair in the dict is the name\n",
    "        of a feature, and its value for that person\n",
    "\n",
    "    In addition to converting a dictionary to a numpy \n",
    "    array, you may want to separate the labels from the\n",
    "    features--this is what targetFeatureSplit is for\n",
    "\n",
    "    so, if you want to have the poi label as the target,\n",
    "    and the features you want to use are the person's\n",
    "    salary and bonus, here's what you would do:\n",
    "\n",
    "    feature_list = [\"poi\", \"salary\", \"bonus\"] \n",
    "    data_array = featureFormat( data_dictionary, feature_list )\n",
    "    label, features = targetFeatureSplit(data_array)\n",
    "\n",
    "    the line above (targetFeatureSplit) assumes that the\n",
    "    label is the _first_ item in feature_list--very important\n",
    "    that poi is listed first!\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "def featureFormat( dictionary, features, remove_NaN=True, remove_all_zeroes=True, remove_any_zeroes=False, sort_keys = False):\n",
    "    \"\"\" convert dictionary to numpy array of features\n",
    "        remove_NaN = True will convert \"NaN\" string to 0.0\n",
    "        remove_all_zeroes = True will omit any data points for which\n",
    "            all the features you seek are 0.0\n",
    "        remove_any_zeroes = True will omit any data points for which\n",
    "            any of the features you seek are 0.0\n",
    "        sort_keys = True sorts keys by alphabetical order. Setting the value as\n",
    "            a string opens the corresponding pickle file with a preset key\n",
    "            order (this is used for Python 3 compatibility, and sort_keys\n",
    "            should be left as False for the course mini-projects).\n",
    "        NOTE: first feature is assumed to be 'poi' and is not checked for\n",
    "            removal for zero or missing values.\n",
    "    \"\"\"\n",
    "\n",
    "\n",
    "    return_list = []\n",
    "\n",
    "    # Key order - first branch is for Python 3 compatibility on mini-projects,\n",
    "    # second branch is for compatibility on final project.\n",
    "    if isinstance(sort_keys, str):\n",
    "        import pickle\n",
    "        keys = pickle.load(open(sort_keys, \"rb\"))\n",
    "    elif sort_keys:\n",
    "        keys = sorted(dictionary.keys())\n",
    "    else:\n",
    "        keys = dictionary.keys()\n",
    "\n",
    "    for key in keys:\n",
    "        tmp_list = []\n",
    "        for feature in features:\n",
    "            try:\n",
    "                dictionary[key][feature]\n",
    "            except KeyError:\n",
    "                print \"error: key \", feature, \" not present\"\n",
    "                return\n",
    "            value = dictionary[key][feature]\n",
    "            if value==\"NaN\" and remove_NaN:\n",
    "                value = 0\n",
    "            tmp_list.append( float(value) )\n",
    "\n",
    "        # Logic for deciding whether or not to add the data point.\n",
    "        append = True\n",
    "        # exclude 'poi' class as criteria.\n",
    "        if features[0] == 'poi':\n",
    "            test_list = tmp_list[1:]\n",
    "        else:\n",
    "            test_list = tmp_list\n",
    "        ### if all features are zero and you want to remove\n",
    "        ### data points that are all zero, do that here\n",
    "        if remove_all_zeroes:\n",
    "            append = False\n",
    "            for item in test_list:\n",
    "                if item != 0 and item != \"NaN\":\n",
    "                    append = True\n",
    "                    break\n",
    "        ### if any features for a given data point are zero\n",
    "        ### and you want to remove data points with any zeroes,\n",
    "        ### handle that here\n",
    "        if remove_any_zeroes:\n",
    "            if 0 in test_list or \"NaN\" in test_list:\n",
    "                append = False\n",
    "        ### Append the data point if flagged for addition.\n",
    "        if append:\n",
    "            return_list.append( np.array(tmp_list) )\n",
    "\n",
    "    return np.array(return_list)\n",
    "\n",
    "\n",
    "def targetFeatureSplit( data ):\n",
    "    \"\"\" \n",
    "        given a numpy array like the one returned from\n",
    "        featureFormat, separate out the first feature\n",
    "        and put it into its own list (this should be the \n",
    "        quantity you want to predict)\n",
    "\n",
    "        return targets and features as separate lists\n",
    "\n",
    "        (sklearn can generally handle both lists and numpy arrays as \n",
    "        input formats when training/predicting)\n",
    "    \"\"\"\n",
    "\n",
    "    target = []\n",
    "    features = []\n",
    "    for item in data:\n",
    "        target.append( item[0] )\n",
    "        features.append( item[1:] )\n",
    "\n",
    "    return target, features\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21.0\n",
      "146.0\n",
      "14.3835616438\n"
     ]
    }
   ],
   "source": [
    "# How many people in the E+F dataset (as it currently exists) have “NaN” for their total payments? \n",
    "# What percentage of people in the dataset as a whole is this?\n",
    "count1 = 0 \n",
    "count2 = 0\n",
    "for p in enron_data:\n",
    "    count2 +=1\n",
    "    for i in enron_data[p]:\n",
    "        if i == \"total_payments\":\n",
    "            temp1 = enron_data[p][i]\n",
    "            #print temp\n",
    "            if temp1 =='NaN':\n",
    "                count1 +=1\n",
    "            else:\n",
    "                pass\n",
    "count1 = float(count1)\n",
    "count2 = float(count2)\n",
    "print count1\n",
    "print count2\n",
    "print (count1/count2)*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "288682\n",
      "1490344\n",
      "1099100\n",
      "2652612\n",
      "2003885\n",
      "4747979\n",
      "103559793\n",
      "2669589\n",
      "5501630\n",
      "2424083\n",
      "1639297\n",
      "505050\n",
      "8682716\n",
      "360300\n",
      "91093\n",
      "1587421\n",
      "1868758\n",
      "1272284\n",
      "18.0\n",
      "0.0\n"
     ]
    }
   ],
   "source": [
    "count1 = 0 \n",
    "count2 = 0\n",
    "for p in enron_data:\n",
    "    for i in enron_data[p]:\n",
    "        if i == \"poi\":\n",
    "            temp1 = enron_data[p][i]\n",
    "            if temp1 == 1:\n",
    "                count1 +=1\n",
    "                print enron_data[p]['total_payments'] \n",
    "                if enron_data[p]['total_payments'] == 'NaN':\n",
    "                    enron_data[p]['total_payments']\n",
    "                    count2 +=1\n",
    "            else:\n",
    "                pass\n",
    "count1 = float(count1)\n",
    "count2 = float(count2)\n",
    "print count1\n",
    "print count2\n",
    "#print (count1/count2)*100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regressions - Continuous Supervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Age/Net Worth Regression in sklearn\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# student Regression.py\n",
    "def studentReg(ages_train, net_worths_train):\n",
    "    ### import the sklearn regression module, create, and train your regression\n",
    "    ### name your regression reg\n",
    "    \n",
    "    ### your code goes here!\n",
    "    from sklearn import linear_model\n",
    "    reg = linear_model.LinearRegression()\n",
    "    reg.fit(ages_train,net_worths_train)\n",
    "    \n",
    "    \n",
    "    return reg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "ename": "ImportError",
     "evalue": "No module named ages_net_worths",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mImportError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-21-cd23a7a08971>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[0mclass_vis\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mprettyPicture\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0moutput_image\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m \u001b[1;32mfrom\u001b[0m \u001b[0mages_net_worths\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mageNetWorthData\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     11\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m \u001b[0mages_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mages_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnet_worths_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnet_worths_test\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mageNetWorthData\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mImportError\u001b[0m: No module named ages_net_worths"
     ]
    }
   ],
   "source": [
    "# StudentMain.py\n",
    "import numpy\n",
    "import matplotlib\n",
    "matplotlib.use('agg')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "#from studentRegression import studentReg\n",
    "from class_vis import prettyPicture, output_image\n",
    "\n",
    "from ages_net_worths import ageNetWorthData\n",
    "\n",
    "ages_train, ages_test, net_worths_train, net_worths_test = ageNetWorthData()\n",
    "\n",
    "\n",
    "\n",
    "reg = studentReg(ages_train, net_worths_train)\n",
    "\n",
    "\n",
    "plt.clf()\n",
    "plt.scatter(ages_train, net_worths_train, color=\"b\", label=\"train data\")\n",
    "plt.scatter(ages_test, net_worths_test, color=\"r\", label=\"test data\")\n",
    "plt.plot(ages_test, reg.predict(ages_test), color=\"black\")\n",
    "plt.legend(loc=2)\n",
    "plt.xlabel(\"ages\")\n",
    "plt.ylabel(\"net worths\")\n",
    "\n",
    "\n",
    "plt.savefig(\"test.png\")\n",
    "output_image(\"test.png\", \"png\", open(\"test.png\", \"rb\").read())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 5.44814029]\n",
      "-102360.543294\n",
      "r-squared score with training data is:  0.0455091926995\n",
      "r-squared score with test data is:  -1.48499241737\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAEKCAYAAABQRFHsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xb8e07f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Regression Mini Project\n",
    "#!/usr/bin/python\n",
    "\n",
    "\"\"\"\n",
    "    Starter code for the regression mini-project.\n",
    "    \n",
    "    Loads up/formats a modified version of the dataset\n",
    "    (why modified?  we've removed some trouble points\n",
    "    that you'll find yourself in the outliers mini-project).\n",
    "\n",
    "    Draws a little scatterplot of the training/testing data\n",
    "\n",
    "    You fill in the regression code where indicated:\n",
    "\"\"\"    \n",
    "\n",
    "\n",
    "import sys\n",
    "import pickle\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "dictionary = pickle.load( open(\"final_project_dataset_modified.pkl\", \"r\") )\n",
    "\n",
    "### list the features you want to look at--first item in the \n",
    "### list will be the \"target\" feature\n",
    "# Regression Bonus against Salary\n",
    "features_list = [\"bonus\", \"salary\"]\n",
    "\n",
    "# Regressing Bonus against LTI\n",
    "#features_list = [\"bonus\", \"long_term_incentive\"]\n",
    "data = featureFormat( dictionary, features_list, remove_any_zeroes=True)\n",
    "target, features = targetFeatureSplit( data )\n",
    "\n",
    "### training-testing split needed in regression, just like classification\n",
    "from sklearn.cross_validation import train_test_split\n",
    "feature_train, feature_test, target_train, target_test = train_test_split(features, target, test_size=0.5, random_state=42)\n",
    "train_color = \"b\"\n",
    "test_color = \"r\"\n",
    "\n",
    "\n",
    "\n",
    "### Your regression goes here!\n",
    "### Please name it reg, so that the plotting code below picks it up and \n",
    "### plots it correctly. Don't forget to change the test_color above from \"b\" to\n",
    "### \"r\" to differentiate training points from test points.\n",
    "\n",
    "from sklearn import linear_model\n",
    "reg = linear_model.LinearRegression()\n",
    "reg.fit(feature_train,target_train)\n",
    "\n",
    "# Extracting Slope and Intercept\n",
    "print reg.coef_\n",
    "print reg.intercept_\n",
    "\n",
    "# Regression Score: Training Data\n",
    "print \"r-squared score with training data is: \", reg.score(feature_train,target_train)\n",
    "\n",
    "# Regression Score: Test Data\n",
    "print \"r-squared score with test data is: \", reg.score(feature_test,target_test)\n",
    "\n",
    "\n",
    "### draw the scatterplot, with color-coded training and testing points\n",
    "import matplotlib.pyplot as plt\n",
    "for feature, target in zip(feature_test, target_test):\n",
    "    plt.scatter( feature, target, color=test_color ) \n",
    "for feature, target in zip(feature_train, target_train):\n",
    "    plt.scatter( feature, target, color=train_color ) \n",
    "\n",
    "### labels for the legend\n",
    "plt.scatter(feature_test[0], target_test[0], color=test_color, label=\"test\")\n",
    "plt.scatter(feature_test[0], target_test[0], color=train_color, label=\"train\")\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### draw the regression line, once it's coded\n",
    "try:\n",
    "    plt.plot( feature_test, reg.predict(feature_test) )\n",
    "except NameError:\n",
    "    pass\n",
    "\n",
    "plt.xlabel(features_list[1])\n",
    "plt.ylabel(features_list[0])\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 2.27410114]\n",
      "124444.388866\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAEKCAYAAABQRFHsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa6153c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sneak Peak: Outliers Break Regressions\n",
    "try:\n",
    "    plt.plot( feature_test, reg.predict(feature_test) )\n",
    "except NameError:\n",
    "    pass\n",
    "reg.fit(feature_test, target_test)\n",
    "print reg.coef_\n",
    "plt.plot(feature_train, reg.predict(feature_train), color=\"b\") \n",
    "plt.xlabel(features_list[1])\n",
    "plt.ylabel(features_list[0])\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#!/usr/bin/python\n",
    "\n",
    "\n",
    "def outlierCleaner(predictions, ages, net_worths):\n",
    "    \"\"\"\n",
    "        Clean away the 10% of points that have the largest\n",
    "        residual errors (difference between the prediction\n",
    "        and the actual net worth).\n",
    "\n",
    "        Return a list of tuples named cleaned_data where \n",
    "        each tuple is of the form (age, net_worth, error).\n",
    "    \"\"\"\n",
    "    from operator import itemgetter\n",
    "    ### your code goes here\n",
    "    error = (predictions-net_worths)**2\n",
    "    cleaned_data = []\n",
    "    cleaned_data = zip(ages,net_worths,error)\n",
    "    cleaned_data = sorted(cleaned_data,key=itemgetter(2),reverse=True)\n",
    "    limit = int(len(cleaned_data)*.1)\n",
    "    cleaned_data = cleaned_data[limit:]\n",
    "    return cleaned_data\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regression Slope is:  [[ 5.07793064]]\n",
      "r-squared score with test data is:  0.878262470366\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHxlJREFUeJzt3X+QHPV55/H3o2WBFY4tcciytAjEnUGUZBmE18CxrpTB\nsYUjYwQ+Y6hwJV+5wj84BXZOkZSSC2zsYonqqKQqVKooEhcOQSDACBklKIDIXc6AdKsflLyADDYI\naUE/bCE5sgVedp/7Y3q0s7PdPTM9PTM9PZ9XFbUz3dMzvW352e8836efr7k7IiKSX1NafQIiItJY\nCvQiIjmnQC8iknMK9CIiOadALyKScwr0IiI5p0AvIpJzCvQiIjmnQC8iknMntfoEAM444wyfO3du\nq09DRKStbNu27VfuPqPS6zIR6OfOncvg4GCrT0NEpK2Y2Z5qXqfUjYhIzinQi4jknAK9iEjOKdCL\niOScAr2ISM4p0IuI5JwCvYhIzmWijl5EpN2s3zHMmk27efvIcWZP62H54nksXdSb+jFpUKAXEanR\n+h3DrPrxLo6PjAIwfOQ4q368CyAycCc5Ji1K3YiI1GjNpt0nAnbR8ZFR1mzaneoxaVGgFxGp0dtH\njte0PekxaVGgFxGp0expPTVtT3pMWhToRaSjrd8xTP/AZs5ZuZH+gc2s3zFc8Zjli+fR0901YVtP\ndxfLF89L9Zi0aDJWRDpW0gnS4r5aKmiSHJMWc/eGf0glfX19rjbFItJs/QObGQ7JkfdO6+GnK69o\nwRnVxsy2uXtfpdcpdSMiHauVE6TNpNSNiHSs2dN6Qkf01UyQturmpyQ0oheRjpV0grSY2x8+chxn\nPLdfzURuK1Qd6M2sy8x2mNmTwfPTzexpM3st+Dm95LWrzOx1M9ttZosbceIiIvVauqiXO69dSO+0\nHoxCbv7OaxdWHJm38uanJGpJ3dwCvAJ8OHi+EnjW3QfMbGXwfIWZzQeuBxYAs4FnzOw8dx8Ne1MR\nkVZauqi35pRLu+X2qxrRm9mZwBLgvpLNVwP3B4/vB5aWbH/I3d939zeA14GL0zldEZHWS3rzU5Ka\n/TRUm7r5a+AvgLGSbTPd/Z3g8X5gZvC4F9hb8rp9wTYRkVxIkttvZV6/YqA3sy8BB919W9RrvFCM\nX1NBvpndZGaDZjZ46NChWg4VEWmpJLn9Vub1q8nR9wNfNrM/Bk4FPmxmDwAHzGyWu79jZrOAg8Hr\nh4E5JcefGWybwN3vBe6Fwg1TdfwOIiJNV2tuP9NNzdx9lbuf6e5zKUyybnb3G4ENwLLgZcuAJ4LH\nG4DrzewUMzsHOBfYmvqZi4g0WJKcetQxrWxqVs8NUwPAOjP7BrAHuA7A3YfMbB3wMvABcLMqbkSk\nVq2+ISntxUWWL543YR80r6mZet2ISOaUB0woBMVqatzTkqQPTqVj0v7jVW2vG7VAEJHMiZu4bFag\nb8TiIklq9tOgFggikjlZuCGp3RYXiaNALyKZUylgNuPGo6SLi3RPsQnbuqdYU/LwcRToRSRz4oJs\ns248StoHB6vwvAWUoxeRzIlbjal/YHOi/H2SidBac+prNu1mZHRigcvIqDd1biGMAr2INFyaQTZJ\n/j6u7BHSW94vC3MLYRToRaShkq7LGiXJYiFRVTy3bxji/Q/GWnpuzaAcvYhULckkaNo9XpYvnkd3\nV9mEZ1f8hGfUiPrI8ZHUzy3JQiaNphG9iFQl6ci8IemM8vs8K9z3GTXSjpL03OLmFlpJgV5EqpL0\nJqa00xlrNu1mZKxswnMsfsIzqv3Aqd1TePd3I6mdG7Tupqg4St2ISFWSjszTTmckOY+oUsnbrlqQ\nyVRL2jSiF5GqJB2Zp53OqOc8oj4za6mWtCnQi0hV6um+mGY6I+0ukFlMtaRNgV5EqpKVicasnEc7\nUZtiEZE2pTbFIpJbrV6UJIl//mdYsmT8eTPH2Ar0ItJW0r7TtpF+/Ws444xWn4XKK0WkzaR9p20j\nmBX+Cwvyv/99c0fzoBG9iGRUVHomq43DLKYd8dAQzJ/fvHMpp0AvIpkTl56Jq6Nvdu7+ttvge98L\n3zdzJuzf37CProkCvYhM0urJzrj0TFQd/eXnz2hK7v611+C886L3Z6CQcRLl6EVkgmat4BQnLj0T\n1c7guVcPNTR3X8y7hwX5o0cLAT6LQR40oheRMkmbl6Vp2tTu0GZj06Z2A+F3s37r4Z2h71VP7j4u\n7/6d70SnbbJGgV5EJsjCZGfUyDhuxJxWl8yvfQ3Wrav93LJMqRsRmSAqMDZzlaSjxyeP5uO2Q31d\nMn/+8/HUTFiQL6Zl2jHIgwK9iJTJwipJSf7YROXu49JNxeA+L+RX27OnvYN7KaVuRHIgzSqZLDQN\nS9qhsppOlHF59+uvh7VrazrVtqBAL9LmGtESICpgrl6/i7Vb9jLqTpcZN1wyh+8vXZj85GM+H9L7\nY7NkSaHXTJQ8jNrjKNCLtLlmVcmsXr+LB15868TzUfcTzxsV7Os5/7ffht6Yw/Me3Esp0Iu0uUZU\nyYSlgtZu2Rv62rVb9jYk0CcVl5rZtg0uuqh555IVCvQibSIqD5/24ttRqaDRiCFw1PZmigvuM2bA\nwYPNO5csUtWNSAus3zFM/8Bmzlm5kf6BzRXvOo27WzXtKpmoVFCUrrgo20DLlo1XzYQpVsx0epAH\nBXqRpkvSYqBSHv4rn+o9EXC7zPjKp5Lnt2tN+dxwyZxEn5PEgQPjwf1HP5q8v93r3RtFgV6kyZL0\nU4/Lw6/fMczDW/eeSKGMuvPw1r1VfUsI+1YRlfLpndbDjZeeNeEPyo2XntWU/HwxuH/sY5P3Pf20\ngnslytGLNFmSydO4PPztG4YYGZsY5UbGnNs3DEWO6uNKMuNq2Jcu6m3axGuljJACe/U0ohdpsiR3\nfcbl4Y9EtAWI2g6VU0G13mGalosuqi7vriBfG43oRZosyV2fcTcQ3RrRtTFOpW8V9daw1+Ldd+H0\n06P3j41VHt1LPAV6kSZLetdnVPCdHtHSd3rQ0jdM2iWZScQF79Wr4Y47mnYquadAL9ICaY6Yb7tq\nAd9et5PSNP0UK2yPkrSXTL2Ud2+Nijl6MzvVzLaa2UtmNmRm3w22n25mT5vZa8HP6SXHrDKz181s\nt5ktbuQvIJI3tdbYA3RNsdjn5ZqZh1+yRHn3VjOvcHXNzIDT3P2YmXUD/xe4BbgWOOzuA2a2Epju\n7ivMbD6wFrgYmA08A5zn7pF3XPT19fng4GA6v5FIGyuvhoHCSDsuCPcPbA5Nw/RO6+GnK69o2LnG\nee896InJAr3/Ppx8cvPOJ6/MbJu791V6XcURvRccC552B/85cDVwf7D9fmBp8Phq4CF3f9/d3wBe\npxD0RaSCtGvsm604cg8L8l/96vjIXUG+uarK0ZtZF7AN+Dhwj7tvMbOZ7v5O8JL9wMzgcS/wYsnh\n+4Jt5e95E3ATwFlnnZXs7EVyJu0a+zhp9bBX3j37qqqjd/dRd78QOBO42Mw+UbbfKYzyq+bu97p7\nn7v3zZgxo5ZDRdperXelJq2xj/v8WtswlPrmN5V3byc13TDl7keA54ArgQNmNgsg+FlsHTQMlDa/\nODPYJiKk36AsycRqkhTR6Oh4cL/nnsn7jx5VcM+qiqkbM5sBjLj7ETPrAT4P3AVsAJYBA8HPJ4JD\nNgAPmtndFCZjzwW2NuDcRdpSXJD96corGNxzeMIqTtU0KKu1XLOWFFFcambRIti+veqPlRapJkc/\nC7g/yNNPAda5+5Nm9gKwzsy+AewBrgNw9yEzWwe8DHwA3BxXcSPSaSo1KHts2/CEBmWPbRum7+zT\nUy19rJTXV949XyqWVzaDyiulk8SVQwJNKZUMK+P8j/89n8MvnhN5TAZChZRJrbxSRNIVl4dvVqlk\nMa8/+yM97LlrCXvuWhIa5PfuVd49D9QCQaTJ4nrdrNm0uyk9aAqpmV5CKp8BBfa8UaAXaZC4OvWo\nydNG9qBR3r1zKdCLVCEuaIftAyIX9oibVE3a2TLK3/4t/NmfRe9XcO8MmowVqSCu/wwQuu/U7imh\nrYOb1X8mbvT+s5/BgujGltJGqp2M1YhepIJKNxeF7SvfVtSo/jPrdwxzzUXxo/4MjOmkRRToRSpI\nsxKmMZOqoElViaNAL1JBpZuLwvZN7Z7C70bGJm2//Pz6+zo9+CD8yZ9E7z97xUagWJffmjbFki0K\n9CIVVKqECdt38knhgf65Vw8lPo+4vPtH/9tWev7LxPduRZtiySYFepEKqqmEKd/3rYgFu2sNvtWU\nREbdadvM9V8l21R1I9IA9az6VCm4z12xccIfm0pVQWmVakr2qOpGJEW1LtKxfPE8lj/yEiMlK3Z3\nT7HIG5+efx76+6M///Ht48G8tLUxRH/jgGS1/JI/CvQiFZSPmKsNmOUZ+skZ+/jR+w9/CF//euFx\n/0B0iWfxLtvyc+kf2Bx7jHQOBXrJvLSWvEsqro4+6jy++5MhRscmpkVHx5zv/mQoUb17khLPLK0l\nK62lQC+ZlnQ0naYkAbP8rtg9dy0p/Ix4faWpsiTrwiZdS1byR22KJROi1lBNsuRd0s+KkmQdV4CR\nw1NPtAAOU8u6qkmWGExyjOSTRvTScut3DLP80ZcYGS1EvOEjx1n+6EtA+umHJN8Qau0oWci7hwf3\n2YtfZfip82s+7yTNztJukCbtS+WV0nKLvvevoQ3Apk/tZurJJ6W64lLSssdK8wSVSiLPXrGR7inG\nmq9eoEArqVF5pTRUmhOkYUG+uP22qxak2p896TeEsKqWmTPh4MHoYx7f3tpJZJEiBXqpWTMnSNNO\nP9Q7QfmrX8GMmHY1Y2MTG40psEsWKNBLzZKUG8aZ1tPNkeOTR/XTerqB6NWYkki6glNcamb1arjj\njlROT6QhFOilZmlPkN7+5QV8++GdE24omhJsT1st3xC09J7khQK91KwR9dldXcbYqE94Xo0kcwVx\n3xA+9SnYvj36WAV3aUeqo5eapV2fvWbT7hOllUUjo16xVr44VzB85PiE/i+V6uLL/fa3hdG7WXiQ\nHxmpvt5dJIsU6KVmSxf1cue1C+md1oNRKE2889qFifPoSVNB9d5MVQzuH/rQ5H1f+cp4cD9J33ul\nzemfsCSS5gRp0lRQkj8QyrtLJ9KIXlouaSqo2tYE1103PnoPU0srApF2pEAvLZc0FRT3B+KDD8aD\n+yOPTD722DEFd+kcaoEgba286ub5VdFtDD75SXjppSaenEiDqQWCVC1JiWLcMavX72Ltlr2MutNl\nxg2XzOH7Sxc25PyWLuo90d/9zYjjMzCWEWkpjehzptagHbfeaNRxcccM7jnMAy++NemYGy89K1Gw\nj/qsj//iMzy5NqRcJpCBf9YiDacRfQdK0oMmSTuDuGP2H30v9Ji1W/YmCvSln+UOb/1Vof3vqyGv\nPXAAPvrRmj9CJPcU6HMkSdBOe4m6qIH0aMIh9ttHjkcu3FGk0btIPFXd5EiSoJ1k9aS4Y7oiahij\ntkcpVsy8GRHkL7tzs6pmRKqkQJ8jSYL25eeH99yN2g7xZY03XDIn9Jio7aXuuSe+3v3sFRs5e8VG\nzl/9lJbDE6mBUjc5kqQF73OvHqppO8R3gCzuq6XqJm6w//Ofw9Cx4gQzWsBDJAFV3eRMrVU3c1du\njNz35kB8brweakUgUj9V3XSoWnvQdJmFTpTWmlOvhoK7SGtUzNGb2Rwze87MXjazITO7Jdh+upk9\nbWavBT+nlxyzysxeN7PdZra4kb+A1CeqGiZplUy5n/xEfWZEWq2aEf0HwJ+7+3Yz+wNgm5k9DXwd\neNbdB8xsJbASWGFm84HrgQXAbOAZMzvP3Ucj3l9aaPrU7tDFuadPLSzjl3QR8LjR+9at8OlPJz5l\nEalRxUDv7u8A7wSP/8PMXgF6gauBzwYvux/4N2BFsP0hd38feMPMXgcuBl5I++SlflEjaffab8BS\nakYkm2oqrzSzucAiYAswM/gjALAfmBk87gX2lhy2L9gmGXQ0ZFHu4vZqFvYopmWUmhHJrqoDvZl9\nCHgMuNXdf1O6zwulOzX9X9nMbjKzQTMbPHQoupRPGiuu9j7qRqtfDp2i4C7SRqoK9GbWTSHI/5O7\n/zjYfMDMZgX7ZwEHg+3DQOndMWcG2yZw93vdvc/d+2bMiL45Rxor7uan8j8Ce+5awp67lrD/gf5J\n7/Mv/6LgLpJVFXP0ZmbA3wOvuPvdJbs2AMuAgeDnEyXbHzSzuylMxp4LbE3zpDtd0gnSMHE3PwEn\nWgBHqTWwp3nuIlKdaqpu+oH/Duwys53Btr+kEODXmdk3gD3AdQDuPmRm64CXKVTs3KyKm/Qk6VBZ\nyeCew+w/+h4O7D/6XklwD3+/pKP2Rpy7iFSmO2PbTP/A5tCFtHun9fDTlVfUvOjH6vW7eODFtxj5\n9Wm8fd9nI1+Xxj+TSucuIrXRnbE5Fdehshi0i0bdTzyPCvY/uGYhEL7v3nvhT/+0vvMtP8datotI\nOtS9ss3EVcms3bI3dF/59kolkcUukWkG+eI51rJdRNKhQN9m4qpk4toZ9PRUF9zPXlFoctaIXjdx\n5y4ijaPUTZuJq5L59rqdjJXE+tFjp7Dvnj+KfK+xMfjOE7tC13itpn98mucuIo2jQN9iaZYbdhmM\nObFL791xB6xePf68mLuvZQK3HrV21xSR+qnqpoXKyw2hkMq489qFkcEw6phXv39l7Gdl4H9mEUmZ\nqm7aQJLFvEuPOfDwxbz3ZvRdxcV8e0HjFhERkWxToG+hJOWG+w69z1t3Rwfts5ZvxMqm2KekP68q\nIm1Egb6FZk/rCb2BKKzccLwI5ouT9n344l/wiWv2hL4XMGGCVkQ6j8orW6hSuWG19e6zv/AayxfP\niyyJbESppIi0DwX6Flq6qJc7r11I77QejEIrgPnD/VxzUW9sC+DHtw9z2Z2bTxxTnLxt9LKAItKe\nlLoJ0cwOi0sX9fKlhb10d8ObEa957z045ZSJx4SdT29EKqhXd56KdDSN6MsUyxeHjxzHGe+wuH7H\n8In9/QObOWflRvoHNp/YnkQxLdPdPXnfzTeP93cvDfJxdOepiITRiL5MpeXz6m2z28h1VXXnqYiE\n0Q1TZc5ZuTF0TUQjukqmUpvdhx6CG26I/sxm/E+gBT9E8kc3TAVqDXBxJY+11L27w5SYxNixY3Da\nacVUUGMDsBb8EOlsuc7RV8q3h6llDdWi0u3FvHtYkL/11vG8ezHI13p+SVRKR4lIvuV6RF+pxUDc\naD9qe1ifmedXXYGtij6PqNRMkhYISWjBD5HOlutAHxfgKqUzwgJt6R+BX+w4jQPrLon87Gry7s0K\nwLXcgSsi+ZPr1E1cqiVpOuOai3p5ftUVoUH+0KHx1Ey95xen1hJPlV2KdLZcB/rLzw/v7Hj5+TNq\nGk3HtSLo//zvTgT3M86o7fyWL55Hd9fEN+3ustgAnCSvH3YHblwrZBHJl1ynbp579VDk9mlTu3n3\ndyOT9k2bWrh76brr4JFHot+72AL4wBRj/Y4LkgfN8tF/hW8DSfP6WvBDpHPlOtDHjdo/0jP5dtSR\nd6ey867LsdvC3++C2/+VI8cn/nEYGXNu3zCUKIiu2bSbkbLWkiNjHhu0NbEqIrXKdaD/SE/3pMBc\n3H60ZHvc0nvvvAMf+1jh8dyVk98LOPEZtdbsJwnamlidSDeCiVSW6xx9VLsBM3j77z7HnruWhAb5\n1avHJ1WLQb6SJLnzJJOxmlgd16z7EETaXa4D/ZGyHPyxodnsuWsJO2/7Ar//zamTXn/+6qd4fPsw\nd9wR/n7Tp4Z0Hwu2J6niSRK0NbE6TjeCiVQn16mb2dN62Hf4Pd5a88eRr7nszs0lX/vjA+ZtVy1g\n+aMvMTI6nlfv7jJuu2oB33p4Z+gxcWmYpE3INLFaoPkKKadUXrjcBvrHH4fnV4U3Gvv2Pw7xv25c\nEDyLbkZWbumiXgb3HGbtlr2MutNlxtc+PYeli3pZs2l3oty5gnZymq+QUurpFC1XqZuhITj33EIO\n/tprx7f3fPwAZ978zIml917cdyDR+6/fMcxj24ZPrNg06s5j24ZZv2NYufMW0DWXUkrlRcvNiP6F\nF+Cyy8afz5kDH3zu3zl55m8mvTbpV/u4f0jFNsX62tg86r/f/tJMtWQllZfF9FFuAv2FF8Kll8It\nt8D11xe29Q+MMHxk8muTfrWv9A9JaZjm0zVvX2mnWrKQystq+ig3qZuensKovhjkIf2v9kl704jI\nZGmnWrKQystq+ig3gT5M2qWIWfiHJJIXaadaslB6nJX0UbncpG6iJPlqH5VjU05YJD2NSLW0OpWX\nhfRRmNwH+lpV06degV2kfssXzwtdyKedvyFn9XfKdeomiazm2ETyJguplrRl9XfSiL5MVnNsInmU\nx2/IWfydNKIvo8oaEckbBfoyqqwRkbypGOjN7B/M7KCZ/axk2+lm9rSZvRb8nF6yb5WZvW5mu81s\ncaNOvFGymmMTEUnKvMJK1mb2h8Ax4Efu/olg218Bh919wMxWAtPdfYWZzQfWAhcDs4FngPPcfTTi\n7QHo6+vzwcHB+n8bEZEOYmbb3L2v0usqTsa6+/8xs7llm68GPhs8vh/4N2BFsP0hd38feMPMXqcQ\n9F+o9sSTyGJvCRGRrEiao5/p7u8Ej/cDM4PHvcDektftC7ZNYmY3mdmgmQ0eOhS+iHc1tMqQiEi8\nuidjvZD7ic//hB93r7v3uXvfjBkzEn++6t5FROIlDfQHzGwWQPDzYLB9GJhT8rozg20No7p3EZF4\nSQP9BmBZ8HgZ8ETJ9uvN7BQzOwc4F9ha3ynGU927iEi8asor11KYTJ1nZvvM7BvAAPB5M3sN+KPg\nOe4+BKwDXgaeAm6uVHFTL9W9i4jEq6bq5oaIXZ+LeP0PgB/Uc1K1UEdJEZF4ueh1k8XeEiIiWZGL\nQC/SCXS/iCSlQC/SBrK6Fqm0BzU1E2kDul9E6qFAL9IGdL+I1EOBXqQN6H4RqYcCvUgb0P0iUg9N\nxoq0Ad0vIvVQoBdpE7pfRJJSoBepk+rbJesU6EXqoPp2aQeajBWpg+rbpR0o0IvUQfXt0g4U6EXq\noPp2aQcK9CJ1UH27tANNxorUQfXt0g46OtCrLE7SoPp2ybqODfQqixORTtGxOXqVxYlIp+jYQK+y\nOBHpFB0b6FUWJyKdomMDvcriRKRTdOxkrMriRKRTdGygB5XFiUhztaqku6MDvYhIs7SypLtjc/Qi\nIs3UypJuBXoRkSZoZUm3Ar2ISBO0sqRbgV5EpAlaWdKtyVgRkSZoZUm3Ar2ISJO0qqRbgV4yQS2j\nRRpHgV5aTi2jRRpLk7HScmoZLdJYCvTScmoZLdJYCvTScmoZLdJYCvTScmoZLdJYmoyVllPLaJHG\naligN7Mrgb8BuoD73H2gUZ8l7U8to0UapyGpGzPrAu4BvgjMB24ws/mN+CwREYnXqBz9xcDr7v5L\nd/898BBwdYM+S0REYjQq0PcCe0ue7wu2iYhIk7Ws6sbMbjKzQTMbPHToUKtOQ0Qk9xoV6IeBOSXP\nzwy2neDu97p7n7v3zZgxo0GnISIijQr0/w8418zOMbOTgeuBDQ36LBERidGQ8kp3/8DMvglsolBe\n+Q/uPtSIzxIRkXjm7q0+B8zsELAnhbc6A/hVCu+TB7oWBboOBboO4/J0Lc5294q570wE+rSY2aC7\n97X6PLJA16JA16FA12FcJ14L9boREck5BXoRkZzLW6C/t9UnkCG6FgW6DgW6DuM67lrkKkcvIiKT\n5W1ELyIiZdo20JvZHDN7zsxeNrMhM7sl2H66mT1tZq8FP6e3+lwbycxONbOtZvZScB2+G2zvqOtQ\nZGZdZrbDzJ4MnnfqdXjTzHaZ2U4zGwy2ddy1MLNpZvaomb1qZq+Y2X/txOvQtoEe+AD4c3efD1wK\n3By0Ql4JPOvu5wLPBs/z7H3gCne/ALgQuNLMLqXzrkPRLcArJc879ToAXO7uF5aUEnbitfgb4Cl3\nPx+4gMK/jc67Du6ei/+AJ4DPA7uBWcG2WcDuVp9bE6/BVGA7cEknXgcKPZWeBa4Angy2ddx1CH7X\nN4EzyrZ11LUAPgK8QTAX2anXwd3bekR/gpnNBRYBW4CZ7v5OsGs/MLNFp9U0QbpiJ3AQeNrdO/I6\nAH8N/AUwVrKtE68DgAPPmNk2M7sp2NZp1+Ic4BDwwyCdd5+ZnUbnXYf2D/Rm9iHgMeBWd/9N6T4v\n/MnOfVmRu4+6+4UURrQXm9knyvbn/jqY2ZeAg+6+Leo1nXAdSnwm+DfxRQppzT8s3dkh1+Ik4CLg\n79x9EfBbytI0HXId2jvQm1k3hSD/T+7+42DzATObFeyfRWGU2xHc/QjwHHAlnXcd+oEvm9mbFFY0\nu8LMHqDzrgMA7j4c/DwIPE5h1bdOuxb7gH3BN1yARykE/k67Du0b6M3MgL8HXnH3u0t2bQCWBY+X\nUcjd55aZzTCzacHjHgrzFK/SYdfB3Ve5+5nuPpdCW+zN7n4jHXYdAMzsNDP7g+Jj4AvAz+iwa+Hu\n+4G9ZjYv2PQ54GU67DpAG98wZWafAf4d2MV4TvYvKeTp1wFnUeiIeZ27H27JSTaBmX0SuJ9CO+gp\nwDp3/56Z/Sc66DqUMrPPAv/T3b/UidfBzP4zhVE8FNIXD7r7Dzr0WlwI3AecDPwS+B8E/z+hk65D\nuwZ6ERGpTtumbkREpDoK9CIiOadALyKScwr0IiI5p0AvIpJzCvQiIjmnQC8iknMK9CIiOff/AbSj\nDHSxvIgbAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa6e97b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!/usr/bin/python\n",
    "# Outliers Mini Project File\n",
    "\n",
    "import random\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "\n",
    "\n",
    "### load up some practice data with outliers in it\n",
    "ages = pickle.load( open(\"practice_outliers_ages.pkl\", \"r\") )\n",
    "net_worths = pickle.load( open(\"practice_outliers_net_worths.pkl\", \"r\") )\n",
    "\n",
    "\n",
    "\n",
    "### ages and net_worths need to be reshaped into 2D numpy arrays\n",
    "### second argument of reshape command is a tuple of integers: (n_rows, n_columns)\n",
    "### by convention, n_rows is the number of data points\n",
    "### and n_columns is the number of features\n",
    "ages       = numpy.reshape( numpy.array(ages), (len(ages), 1))\n",
    "net_worths = numpy.reshape( numpy.array(net_worths), (len(net_worths), 1))\n",
    "from sklearn.cross_validation import train_test_split\n",
    "ages_train, ages_test, net_worths_train, net_worths_test = train_test_split(ages, net_worths, test_size=0.1, random_state=42)\n",
    "\n",
    "### fill in a regression here!  Name the regression object reg so that\n",
    "### the plotting code below works, and you can see what your regression looks like\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "reg = LinearRegression()\n",
    "reg.fit(ages_train,net_worths_train)\n",
    "\n",
    "# Slope of Regression With Outliers\n",
    "print \"Regression Slope is: \", reg.coef_\n",
    "\n",
    "# Score of Regressions with Outliers\n",
    "print \"r-squared score with test data is: \", reg.score(ages_test,net_worths_test)\n",
    "\n",
    "\n",
    "try:\n",
    "    plt.plot(ages, reg.predict(ages), color=\"blue\")\n",
    "except NameError:\n",
    "    pass\n",
    "plt.scatter(ages, net_worths)\n",
    "plt.show()\n",
    "\n",
    "\n",
    "### identify and remove the most outlier-y points\n",
    "cleaned_data = []\n",
    "\n",
    "#print \"printting predictions\"\n",
    "predictions = reg.predict(ages_train)\n",
    "#print predictions\n",
    "#print ages\n",
    "\n",
    "cleaned_data = outlierCleaner( predictions, ages_train, net_worths_train )\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEKCAYAAAAIO8L1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc4a5518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regression Slope is:  [[ 6.36859481]]\n",
      "r-squared score with test data is:  0.983189455396\n"
     ]
    }
   ],
   "source": [
    "### only run this code if cleaned_data is returning data\n",
    "if len(cleaned_data) > 0:\n",
    "    ages, net_worths, errors = zip(*cleaned_data)\n",
    "    ages       = numpy.reshape( numpy.array(ages), (len(ages), 1))\n",
    "    net_worths = numpy.reshape( numpy.array(net_worths), (len(net_worths), 1))\n",
    "\n",
    "    ### refit your cleaned data!\n",
    "    try:\n",
    "        reg.fit(ages, net_worths)\n",
    "        plt.plot(ages, reg.predict(ages), color=\"blue\")\n",
    "    except NameError:\n",
    "        print \"you don't seem to have regression imported/created,\"\n",
    "        print \"   or else your regression object isn't named reg\"\n",
    "        print \"   either way, only draw the scatter plot of the cleaned data\"\n",
    "    plt.scatter(ages, net_worths)\n",
    "    plt.xlabel(\"ages\")\n",
    "    plt.ylabel(\"net worths\")\n",
    "    plt.show()\n",
    "    # Slope of Regression Without Outliers\n",
    "    print \"Regression Slope is: \", reg.coef_\n",
    "\n",
    "    # Score of Regressions without Outliers\n",
    "    print \"r-squared score with test data is: \", reg.score(ages_test,net_worths_test)\n",
    "\n",
    "\n",
    "else:\n",
    "    print \"outlierCleaner() is returning an empty list, no refitting to be done\"\n",
    "    \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAERCAYAAACU1LsdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSVLBUJAkFQwFSVLBUJAkFQwFSVLh/wP1B+q3c2iFtAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc9fca20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Enron Outliers\n",
    "#!/usr/bin/python\n",
    "\n",
    "import pickle\n",
    "import sys\n",
    "import matplotlib.pyplot\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "### read in data dictionary, convert to numpy array\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "features = [\"salary\", \"bonus\"]\n",
    "data = featureFormat(data_dict, features)\n",
    "\n",
    "\n",
    "### your code below\n",
    "\n",
    "\n",
    "for point in data:\n",
    "    salary = point[0]\n",
    "    bonus = point[1]\n",
    "    matplotlib.pyplot.scatter( salary, bonus )\n",
    "\n",
    "matplotlib.pyplot.xlabel(\"salary\")\n",
    "matplotlib.pyplot.ylabel(\"bonus\")\n",
    "matplotlib.pyplot.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAEKCAYAAABQRFHsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEYmVEpGIiMRKiUhERGKlRCQiIrH6/4lw3FIdlKZsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xca6cdd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Any More Outliers?\n",
    "import pickle\n",
    "import sys\n",
    "import matplotlib.pyplot\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "### read in data dictionary, convert to numpy array\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "features = [\"salary\", \"bonus\"]\n",
    "# Removing and checking for more outliers\n",
    "data_dict.pop('TOTAL',0)\n",
    "data = featureFormat(data_dict, features)\n",
    "\n",
    "\n",
    "### your code below\n",
    "\n",
    "\n",
    "for point in data:\n",
    "    salary = point[0]\n",
    "    bonus = point[1]\n",
    "    matplotlib.pyplot.scatter( salary, bonus )\n",
    "\n",
    "matplotlib.pyplot.xlabel(\"salary\")\n",
    "matplotlib.pyplot.ylabel(\"bonus\")\n",
    "matplotlib.pyplot.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Employes with Bonus > 5000000 \n",
      "\n",
      "LAVORATO JOHN J\n",
      "LAY KENNETH L\n",
      "BELDEN TIMOTHY N\n",
      "SKILLING JEFFREY K\n",
      "Employes with Salary > 1000000 \n",
      "\n",
      "LAY KENNETH L\n",
      "SKILLING JEFFREY K\n",
      "FREVERT MARK A\n"
     ]
    }
   ],
   "source": [
    "print \"Employes with Bonus > 5000000 \\n\"\n",
    "for p in data_dict:\n",
    "    if data_dict[p]['bonus'] >= 5000000:\n",
    "        if data_dict[p]['bonus'] !='NaN':\n",
    "            print p\n",
    "            \n",
    "print \"Employes with Salary > 1000000 \\n\"            \n",
    "for p in data_dict:\n",
    "    if data_dict[p]['salary'] >= 1000000:\n",
    "        if data_dict[p]['salary'] !='NaN':\n",
    "            print p\n",
    "        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clustering - K Means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def Draw(pred, features, poi, mark_poi=False, name=\"image.png\", f1_name=\"feature 1\", f2_name=\"feature 2\"):\n",
    "    \"\"\" some plotting code designed to help you visualize your clusters \"\"\"\n",
    "\n",
    "    ### plot each cluster with a different color--add more colors for\n",
    "    ### drawing more than five clusters\n",
    "    colors = [\"b\", \"c\", \"k\", \"m\", \"g\"]\n",
    "    for ii, pp in enumerate(pred):\n",
    "        plt.scatter(features[ii][0], features[ii][1], color = colors[pred[ii]])\n",
    "\n",
    "    ### if you like, place red stars over points that are POIs (just for funsies)\n",
    "    if mark_poi:\n",
    "        for ii, pp in enumerate(pred):\n",
    "            if poi[ii]:\n",
    "                plt.scatter(features[ii][0], features[ii][1], color=\"r\", marker=\"*\")\n",
    "    plt.xlabel(f1_name)\n",
    "    plt.ylabel(f2_name)\n",
    "    plt.savefig(name)\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEDCAYAAADOc0QpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x81820b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAERCAYAAAB/4wAeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBAsSRE4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9ca0d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# k_means_cluster.py\n",
    "#!/usr/bin/python \n",
    "\n",
    "\"\"\" \n",
    "    Skeleton code for k-means clustering mini-project.\n",
    "\"\"\"\n",
    "import pickle\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### load in the dict of dicts containing all the data on each person in the dataset\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "### there's an outlier--remove it! \n",
    "data_dict.pop(\"TOTAL\", 0)\n",
    "\n",
    "\n",
    "### the input features we want to use \n",
    "### can be any key in the person-level dictionary (salary, director_fees, etc.) \n",
    "feature_1 = \"salary\"\n",
    "feature_2 = \"exercised_stock_options\"\n",
    "poi  = \"poi\"\n",
    "features_list = [poi, feature_1, feature_2]\n",
    "data = featureFormat(data_dict, features_list )\n",
    "poi, finance_features = targetFeatureSplit( data )\n",
    "\n",
    "\n",
    "### in the \"clustering with 3 features\" part of the mini-project,\n",
    "### you'll want to change this line to \n",
    "### for f1, f2, _ in finance_features:\n",
    "### (as it's currently written, the line below assumes 2 features)\n",
    "for f1, f2 in finance_features:\n",
    "    plt.scatter( f1, f2 )\n",
    "plt.show()\n",
    "\n",
    "### cluster here; create predictions of the cluster labels\n",
    "### for the data and store them to a list called pred\n",
    "from sklearn.cluster import KMeans\n",
    "kmeans = KMeans(n_clusters=2, random_state=0)\n",
    "kmeans.fit(finance_features)\n",
    "pred = kmeans.predict(finance_features)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### rename the \"name\" parameter when you change the number of features\n",
    "### so that the figure gets saved to a different file\n",
    "try:\n",
    "    Draw(pred, finance_features, poi, mark_poi=False, name=\"clusters.pdf\", f1_name=feature_1, f2_name=feature_2)\n",
    "except NameError:\n",
    "    print \"no predictions object named pred found, no clusters to plot\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "print Finance freatures /n [array([  365788.,        0.,  1061827.]), array([  267102.,  6680544.,  5634343.]), array([  170941.,  4890344.,   211725.]), array([      0.,  651850.,       0.]), array([  243293.,  5538001.,   288682.]), array([ 267093.,       0.,  628522.]), array([       0.,   493489.,  1848227.]), array([  370448.,  1104054.,  4099771.]), array([       0.,  5210569.,  3131860.]), array([  197091.,   880290.,  1737629.]), array([  130724.,  2282768.,  3100224.]), array([  288589.,        0.,  1130461.]), array([  248546.,   765313.,  2014835.]), array([  257486.,        0.,  1545059.]), array([       0.,  2218275.,   966522.]), array([      0.,  372205.,       0.]), array([  288542.,        0.,  1490344.]), array([  251654.,  1056320.,  3202070.]), array([ 288558.,  185063.,  551174.]), array([  63744.,       0.,  762135.]), array([       0.,  1030329.,    87410.]), array([  357091.,  4346544.,  1798780.]), array([  271442.,    81042.,  3471141.]), array([      0.,    3285.,  102500.]), array([  304110.,  1441898.,  3038702.]), array([      0.,       0.,  189583.]), array([ 187922.,       0.,  618850.]), array([      0.,       0.,  228656.]), array([  213625.,  1465734.,  2047593.]), array([  249201.,  1635238.,  1099100.]), array([     0.,  98718.,      0.]), array([ 231330.,       0.,  969068.]), array([   0.,    0.,  475.]), array([  182245.,   601438.,  2692324.]), array([   0.,    0.,  148.]), array([  211788.,        0.,  2081796.]), array([      0.,  850477.,       0.]), array([    0.,     0.,  7961.]), array([      0.,  664375.,       0.]), array([  224305.,        0.,  2652612.]), array([  273746.,        0.,  2664228.]), array([   339288.,   4158995.,  10425757.]), array([  216582.,  1624396.,   228474.]), array([  210500.,  2027865.,  1204583.]), array([      0.,       0.,  213071.]), array([  272880.,   436515.,  2146973.]), array([  4.77000000e+02,   4.04615700e+06,   9.16197000e+05]), array([      0.,  371750.,   77492.]), array([  269076.,   607837.,  1057548.]), array([  428780.,  1835558.,  4335388.]), array([  211844.,  1624396.,  2003885.]), array([      0.,  412878.,   87492.]), array([  206121.,    17378.,  1208649.]), array([ 174246.,       0.,  704896.]), array([  510364.,  3282960.,  4677574.]), array([  365038.,  1623010.,  2101364.]), array([  365163.,  2291113.,  4747979.]), array([  162779.,  1362375.,   999356.]), array([      0.,  139130.,       0.]), array([  236457.,  2549361.,   875889.]), array([    0.,     0.,  2151.]), array([  1.07232100e+06,   3.43483840e+07,   1.03559793e+08]), array([  261516.,        0.,  1271582.]), array([  329078.,  1637034.,  1321557.]), array([      0.,  757301.,       0.]), array([ 184899.,       0.,  807956.]), array([ 192008.,   83237.,  900585.]), array([  263413.,   506765.,  1318763.]), array([  262663.,   187500.,  1297461.]), array([      0.,  759557.,       0.]), array([       0.,  1753766.,        0.]), array([  374125.,   608750.,  3859065.]), array([  278601.,        0.,  2669589.]), array([    0.,  9803.,     0.]), array([  199157.,   664461.,  1414857.]), array([     0.,  59539.,      0.]), array([   96840.,  7509039.,   111529.]), array([   80818.,  1599641.,   860136.]), array([  213999.,   953136.,  5501630.]), array([  210692.,  1451869.,  2093263.]), array([  222093.,  4452476.,   911453.]), array([  440698.,        0.,  2424083.]), array([      0.,  343434.,  181755.]), array([  240189.,        0.,  1639297.]), array([   420636.,  19794175.,    505050.]), array([  275101.,   850010.,  1086821.]), array([  314288.,   117551.,  1101393.]), array([   94941.,  5266578.,   249787.]), array([     0.,      0.,  27942.]), array([  239502.,        0.,  1112087.]), array([  1111258.,  19250000.,   8682716.]), array([    0.,     0.,  1279.]), array([       0.,  1426469.,        0.]), array([      0.,  886231.,  564348.]), array([    6615.,        0.,  1146658.]), array([  655037.,    28798.,  1386690.]), array([      0.,       0.,  362096.]), array([      0.,       0.,  774401.]), array([  404338.,  2022048.,  1747522.]), array([       0.,  1324578.,        0.]), array([  259996.,  1668260.,  1034395.]), array([  317543.,  1297049.,  1934359.]), array([       0.,  5127155.,        0.]), array([  201955.,  1729541.,  4484442.]), array([  248146.,   281073.,  1576511.]), array([      0.,  431750.,   83750.]), array([     0.,      0.,  84992.]), array([     0.,      0.,  49288.]), array([      0.,  257817.,  182466.]), array([      0.,  754966.,       0.]), array([   76399.,  4160672.,   394475.]), array([  262788.,  8191755.,  2368151.]), array([        0.,   2604490.,  15456290.]), array([ 261809.,       0.,  477557.]), array([  248017.,   825464.,  1054637.]), array([      0.,  192758.,   55097.]), array([ 229284.,  400478.,  812194.]), array([  231946.,   636246.,  1566469.]), array([ 221003.,       0.,  319941.]), array([  158403.,  8308552.,   360300.]), array([        0.,  30766064.,     91093.]), array([  250100.,  1550019.,  1410464.]), array([  492375.,  8831913.,   550981.]), array([  1060932.,  10433518.,  17252530.]), array([   261879.,  15364167.,   3123383.]), array([ 239671.,       0.,  827696.]), array([      0.,  176378.,   50591.]), array([  304588.,   940257.,  3676340.]), array([  309946.,   671737.,  1587421.]), array([    0.,     0.,  3750.]), array([   85274.,  2165172.,  1979596.]), array([ 247338.,  591250.,  399393.]), array([  349487.,  2070306.,  8407016.]), array([  330546.,  2542813.,  2355702.]), array([      0.,       0.,  119292.]), array([  415189.,        0.,  1868758.]), array([  265214.,  3181250.,  1092663.]), array([ 278601.,  765920.,  875760.]), array([  274975.,   384728.,  1272284.])]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEDCAYAAADOc0QpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2b78a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAERCAYAAAB/4wAeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xbd2d048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Clustering with 3 features\n",
    "feature_1 = \"salary\"\n",
    "feature_2 = \"exercised_stock_options\"\n",
    "feature_3 = \"total_payments\"\n",
    "poi  = \"poi\"\n",
    "features_list = [poi, feature_1, feature_2, feature_3]\n",
    "data = featureFormat(data_dict, features_list )\n",
    "poi, finance_features = targetFeatureSplit( data )\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### in the \"clustering with 3 features\" part of the mini-project,\n",
    "### you'll want to change this line to \n",
    "### for f1, f2, _ in finance_features:\n",
    "### (as it's currently written, the line below assumes 2 features)\n",
    "for f1, f2, _ in finance_features:\n",
    "    plt.scatter( f1, f2 )\n",
    "plt.show()\n",
    "\n",
    "### cluster here; create predictions of the cluster labels\n",
    "### for the data and store them to a list called pred\n",
    "from sklearn.cluster import KMeans\n",
    "kmeans = KMeans(n_clusters=2, random_state=0)\n",
    "kmeans.fit(finance_features)\n",
    "pred = kmeans.predict(finance_features)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### rename the \"name\" parameter when you change the number of features\n",
    "### so that the figure gets saved to a different file\n",
    "try:\n",
    "    Draw(pred, finance_features, poi, mark_poi=False, name=\"clusters2.pdf\", f1_name=feature_1, f2_name=feature_2)\n",
    "except NameError:\n",
    "    print \"no predictions object named pred found, no clusters to plot\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "101\n",
      "3285\n",
      "34348384\n"
     ]
    }
   ],
   "source": [
    "exercised_stock_options = []\n",
    "\n",
    "for p in data_dict:\n",
    "    if data_dict[p]['exercised_stock_options'] == 'NaN':\n",
    "        pass\n",
    "    else:\n",
    "        exercised_stock_options.append(data_dict[p]['exercised_stock_options'])\n",
    "        \n",
    "    \n",
    "    \n",
    "print len(exercised_stock_options)\n",
    "print min(exercised_stock_options)\n",
    "print max(exercised_stock_options)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "94\n",
      "477\n",
      "1111258\n"
     ]
    }
   ],
   "source": [
    "salaries = []\n",
    "\n",
    "for p in data_dict:\n",
    "    if data_dict[p]['salary'] == 'NaN':\n",
    "        pass\n",
    "    else:\n",
    "        salaries.append(data_dict[p]['salary'])\n",
    "        \n",
    "    \n",
    "    \n",
    "print len(salaries)\n",
    "print min(salaries)\n",
    "print max(salaries)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Feature Scaling\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5\n"
     ]
    }
   ],
   "source": [
    "#Min/Max Rescaler Coding Quiz\n",
    "\"\"\" quiz materials for feature scaling clustering \"\"\"\n",
    "\n",
    "### FYI, the most straightforward implementation might \n",
    "### throw a divide-by-zero error, if the min and max\n",
    "### values are the same\n",
    "### but think about this for a second--that means that every\n",
    "### data point has the same value for that feature!  \n",
    "### why would you rescale it?  Or even use it at all?\n",
    "def featureScaling(arr):\n",
    "    xmin = arr[0]\n",
    "    xmax=arr[2]\n",
    "    x = arr[1]\n",
    "    \n",
    "    xmin = float(xmin)\n",
    "    x = float(x)\n",
    "    xmax = float(xmax)\n",
    "    \n",
    "    if xmin==xmax:\n",
    "        return .5\n",
    "    else:\n",
    "        xprime = (x-xmin)/(xmax-xmin)\n",
    "        return xprime\n",
    "\n",
    "\n",
    "# tests of your feature scaler--line below is input data\n",
    "data = [115, 140, 175]\n",
    "data1 = [15, 15, 15]\n",
    "\n",
    "print featureScaling(data1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.        ],\n",
       "       [ 0.41666667],\n",
       "       [ 1.        ]])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Using Sklearn for Min/Max Scaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "import numpy\n",
    "weights = numpy.array([[115.],[140.],[175.]])\n",
    "scaler = MinMaxScaler()\n",
    "rescaled_weight = scaler.fit_transform(weights)\n",
    "rescaled_weight\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "printing rescaled_features \n",
      "[[  3.29165684e-01   0.00000000e+00]\n",
      " [  2.40360024e-01   1.94493691e-01]\n",
      " [  1.53826564e-01   1.42374791e-01]\n",
      " [  0.00000000e+00   1.89776031e-02]\n",
      " [  2.18934757e-01   1.61230322e-01]\n",
      " [  2.40351925e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   1.43671679e-02]\n",
      " [  3.33359130e-01   3.21428222e-02]\n",
      " [  0.00000000e+00   1.51697646e-01]\n",
      " [  1.77358453e-01   2.56282799e-02]\n",
      " [  1.17636049e-01   6.64592547e-02]\n",
      " [  2.59695768e-01   0.00000000e+00]\n",
      " [  2.23661832e-01   2.22809026e-02]\n",
      " [  2.31706768e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   6.45816409e-02]\n",
      " [  0.00000000e+00   1.08361721e-02]\n",
      " [  2.59653474e-01   0.00000000e+00]\n",
      " [  2.26458662e-01   3.07531207e-02]\n",
      " [  2.59667872e-01   5.38782261e-03]\n",
      " [  5.73620167e-02   0.00000000e+00]\n",
      " [  0.00000000e+00   2.99964330e-02]\n",
      " [  3.21339419e-01   1.26542896e-01]\n",
      " [  2.44265508e-01   2.35941231e-03]\n",
      " [  0.00000000e+00   9.56376871e-05]\n",
      " [  2.73662822e-01   4.19786270e-02]\n",
      " [  1.69107444e-01   0.00000000e+00]\n",
      " [  1.92237086e-01   4.26725752e-02]\n",
      " [  2.24251254e-01   4.76074216e-02]\n",
      " [  0.00000000e+00   2.87402167e-03]\n",
      " [  2.08169480e-01   0.00000000e+00]\n",
      " [  1.63998819e-01   1.75099358e-02]\n",
      " [  1.90584005e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   2.47603206e-02]\n",
      " [  0.00000000e+00   1.93422491e-02]\n",
      " [  2.01847816e-01   0.00000000e+00]\n",
      " [  2.46338834e-01   0.00000000e+00]\n",
      " [  3.05318837e-01   1.21082698e-01]\n",
      " [  1.94898034e-01   4.72917736e-02]\n",
      " [  1.89424958e-01   5.90381486e-02]\n",
      " [  2.45559537e-01   1.27084581e-02]\n",
      " [  4.29243254e-04   1.17797594e-01]\n",
      " [  0.00000000e+00   1.08229255e-02]\n",
      " [  2.42136390e-01   1.76962328e-02]\n",
      " [  3.85850990e-01   5.34394282e-02]\n",
      " [  1.90634398e-01   4.72917736e-02]\n",
      " [  0.00000000e+00   1.20203035e-02]\n",
      " [  1.85484379e-01   5.05933554e-04]\n",
      " [  1.56800671e-01   0.00000000e+00]\n",
      " [  4.59266885e-01   9.55782956e-02]\n",
      " [  3.28490774e-01   4.72514224e-02]\n",
      " [  3.28603259e-01   6.67022064e-02]\n",
      " [  1.46481735e-01   3.96634380e-02]\n",
      " [  0.00000000e+00   4.05055446e-03]\n",
      " [  2.12783170e-01   7.42206970e-02]\n",
      " [  9.64961332e-01   1.00000000e+00]\n",
      " [  2.35333289e-01   0.00000000e+00]\n",
      " [  2.96131051e-01   4.76597094e-02]\n",
      " [  0.00000000e+00   2.20476457e-02]\n",
      " [  1.66387104e-01   0.00000000e+00]\n",
      " [  1.72784358e-01   2.42331633e-03]\n",
      " [  2.37040363e-01   1.47536781e-02]\n",
      " [  2.36365452e-01   5.45877209e-03]\n",
      " [  0.00000000e+00   2.21133256e-02]\n",
      " [  0.00000000e+00   5.10581808e-02]\n",
      " [  3.36667992e-01   1.77228134e-02]\n",
      " [  2.50707756e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   2.85399162e-04]\n",
      " [  1.79217607e-01   1.93447529e-02]\n",
      " [  0.00000000e+00   1.73338577e-03]\n",
      " [  8.71444795e-02   2.18614040e-01]\n",
      " [  7.27265855e-02   4.65710701e-02]\n",
      " [  1.92573642e-01   2.77490784e-02]\n",
      " [  1.89597735e-01   4.22689172e-02]\n",
      " [  1.99857279e-01   1.29626943e-01]\n",
      " [  3.96575773e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   9.99854899e-03]\n",
      " [  2.16141526e-01   0.00000000e+00]\n",
      " [  3.78522359e-01   5.76276747e-01]\n",
      " [  2.47558173e-01   2.47467246e-02]\n",
      " [  2.82821811e-01   3.42231530e-03]\n",
      " [  8.54356054e-02   1.53328261e-01]\n",
      " [  2.15523308e-01   0.00000000e+00]\n",
      " [  1.00000000e+00   5.60433935e-01]\n",
      " [  0.00000000e+00   4.15294356e-02]\n",
      " [  0.00000000e+00   2.58012429e-02]\n",
      " [  5.95271305e-03   0.00000000e+00]\n",
      " [  5.89455374e-01   8.38409166e-04]\n",
      " [  3.63856098e-01   5.88687957e-02]\n",
      " [  0.00000000e+00   3.85630369e-02]\n",
      " [  2.33965470e-01   4.85688060e-02]\n",
      " [  2.85750924e-01   3.77615727e-02]\n",
      " [  0.00000000e+00   1.49269177e-01]\n",
      " [  1.81735475e-01   5.03529074e-02]\n",
      " [  2.23301879e-01   8.18300506e-03]\n",
      " [  0.00000000e+00   1.25697325e-02]\n",
      " [  0.00000000e+00   7.50594264e-03]\n",
      " [  0.00000000e+00   2.19796658e-02]\n",
      " [  6.87500112e-02   1.21131521e-01]\n",
      " [  2.36477938e-01   2.38490259e-01]\n",
      " [  0.00000000e+00   7.58256924e-02]\n",
      " [  2.35596954e-01   0.00000000e+00]\n",
      " [  2.23185795e-01   2.40321058e-02]\n",
      " [  0.00000000e+00   5.61185062e-03]\n",
      " [  2.06328323e-01   1.16592967e-02]\n",
      " [  2.08723807e-01   1.85233168e-02]\n",
      " [  1.98876409e-01   0.00000000e+00]\n",
      " [  1.42543856e-01   2.41890623e-01]\n",
      " [  0.00000000e+00   8.95706302e-01]\n",
      " [  2.25060247e-01   4.51264025e-02]\n",
      " [  4.43078925e-01   2.57127468e-01]\n",
      " [  9.54712587e-01   3.03755717e-01]\n",
      " [  2.35659946e-01   4.47303926e-01]\n",
      " [  2.15675388e-01   0.00000000e+00]\n",
      " [  0.00000000e+00   5.13497229e-03]\n",
      " [  2.74092965e-01   2.73741262e-02]\n",
      " [  2.78914527e-01   1.95565823e-02]\n",
      " [  7.67364554e-02   6.30356293e-02]\n",
      " [  2.22574776e-01   1.72133280e-02]\n",
      " [  3.14496724e-01   6.02737526e-02]\n",
      " [  2.97452077e-01   7.40300621e-02]\n",
      " [  3.73620707e-01   0.00000000e+00]\n",
      " [  2.38661049e-01   9.26171665e-02]\n",
      " [  2.50707756e-01   2.22985745e-02]\n",
      " [  2.47444788e-01   1.12007598e-02]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xba33f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEKCAYAAAAW8vJGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd872908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# k_means_cluster.py\n",
    "#!/usr/bin/python \n",
    "\n",
    "\"\"\" \n",
    "    Skeleton code for k-means clustering mini-project.\n",
    "\"\"\"\n",
    "import pickle\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### load in the dict of dicts containing all the data on each person in the dataset\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "### there's an outlier--remove it! \n",
    "data_dict.pop(\"TOTAL\", 0)\n",
    "\n",
    "\n",
    "### the input features we want to use \n",
    "### can be any key in the person-level dictionary (salary, director_fees, etc.) \n",
    "feature_1 = \"salary\"\n",
    "feature_2 = \"exercised_stock_options\"\n",
    "poi  = \"poi\"\n",
    "features_list = [poi, feature_1, feature_2]\n",
    "data = featureFormat(data_dict, features_list )\n",
    "poi, finance_features = targetFeatureSplit( data )\n",
    "\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "new_finance_features = np.array(finance_features) +0.\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "rescaled_features = scaler.fit_transform(finance_features)\n",
    "\n",
    "\n",
    "print \"printing rescaled_features \\n\", rescaled_features\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### in the \"clustering with 3 features\" part of the mini-project,\n",
    "### you'll want to change this line to \n",
    "### for f1, f2, _ in finance_features:\n",
    "### (as it's currently written, the line below assumes 2 features)\n",
    "for f1, f2 in rescaled_features:\n",
    "    plt.scatter( f1, f2 )\n",
    "plt.show()\n",
    "\n",
    "### cluster here; create predictions of the cluster labels\n",
    "### for the data and store them to a list called pred\n",
    "from sklearn.cluster import KMeans\n",
    "kmeans = KMeans(n_clusters=2, random_state=0)\n",
    "kmeans.fit(rescaled_features)\n",
    "pred = kmeans.predict(rescaled_features)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### rename the \"name\" parameter when you change the number of features\n",
    "### so that the figure gets saved to a different file\n",
    "try:\n",
    "    Draw(pred, rescaled_features, poi, mark_poi=False, name=\"clusters.pdf\", f1_name=feature_1, f2_name=feature_2)\n",
    "except NameError:\n",
    "    print \"no predictions object named pred found, no clusters to plot\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.        ]\n",
      " [ 0.17962407]\n",
      " [ 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "salary = []\n",
    "for p in data_dict:\n",
    "    val = data_dict[p]['salary']\n",
    "    if val =='NaN':\n",
    "        pass\n",
    "    else:\n",
    "        salary.append(float(val))\n",
    "    \n",
    "minSalary = min(salary)\n",
    "maxSalary = max(salary)\n",
    "\n",
    "scaler = MinMaxScaler()      \n",
    "weights = numpy.array([[minSalary],[200000.],[maxSalary]])\n",
    "\n",
    "rescaled_weight = scaler.fit_transform(weights)\n",
    "print rescaled_weight     \n",
    "        \n",
    "        \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.        ]\n",
      " [ 0.02902059]\n",
      " [ 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "ex_stock= []\n",
    "for p in data_dict:\n",
    "    val = data_dict[p]['exercised_stock_options']\n",
    "    if val =='NaN':\n",
    "        pass\n",
    "    else:\n",
    "        ex_stock.append(float(val))\n",
    "    \n",
    "min_ex_stock = min(ex_stock)\n",
    "max_ex_stock = max(ex_stock)\n",
    "\n",
    "scaler = MinMaxScaler()      \n",
    "weights = numpy.array([[min_ex_stock],[1000000.],[max_ex_stock]])\n",
    "\n",
    "rescaled_weight = scaler.fit_transform(weights)\n",
    "print rescaled_weight     \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niVA4iIhIhMJBREQiFA4iIhLx/wH9E4lwchtXowAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x99e0588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZYAAAEKCAYAAAAxXHOuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x99e09b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# When to Deploy Feature Scaling - No feature scaling\n",
    "# k_means_cluster.py\n",
    "#!/usr/bin/python \n",
    "\n",
    "\"\"\" \n",
    "    Skeleton code for k-means clustering mini-project.\n",
    "\"\"\"\n",
    "import pickle\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### load in the dict of dicts containing all the data on each person in the dataset\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "### there's an outlier--remove it! \n",
    "data_dict.pop(\"TOTAL\", 0)\n",
    "\n",
    "\n",
    "### the input features we want to use \n",
    "### can be any key in the person-level dictionary (salary, director_fees, etc.) \n",
    "feature_1 = \"salary\"\n",
    "feature_2 = \"from_messages\"\n",
    "poi  = \"poi\"\n",
    "features_list = [poi, feature_1, feature_2]\n",
    "data = featureFormat(data_dict, features_list )\n",
    "poi, finance_features = targetFeatureSplit( data )\n",
    "\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "new_finance_features = np.array(finance_features) +0.\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "rescaled_features = scaler.fit_transform(finance_features)\n",
    "\n",
    "\n",
    "### in the \"clustering with 3 features\" part of the mini-project,\n",
    "### you'll want to change this line to \n",
    "### for f1, f2, _ in finance_features:\n",
    "### (as it's currently written, the line below assumes 2 features)\n",
    "for f1, f2 in finance_features:\n",
    "    plt.scatter( f1, f2 )\n",
    "plt.show()\n",
    "\n",
    "### cluster here; create predictions of the cluster labels\n",
    "### for the data and store them to a list called pred\n",
    "from sklearn.cluster import KMeans\n",
    "kmeans = KMeans(n_clusters=2, random_state=0)\n",
    "kmeans.fit(finance_features)\n",
    "pred = kmeans.predict(finance_features)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### rename the \"name\" parameter when you change the number of features\n",
    "### so that the figure gets saved to a different file\n",
    "try:\n",
    "    Draw(pred, finance_features, poi, mark_poi=False, name=\"clusters4.jpg\", f1_name=feature_1, f2_name=feature_2)\n",
    "except NameError:\n",
    "    print \"no predictions object named pred found, no clusters to plot\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHcJJREFUeJzt3XtwVOeZ5/Hv0y2pJXMTF2EJgQ0igM0mOHZkfCkT22Fj\nfJkMHjs7i3Mh8Vxcro0T78zi2Iw3iSuprDNrTyXO2gnLpFwZpzJhvIHlMvYM3iKOTXxJkI0DxoAN\nimMkJCMQCBC6dKuf/aNbQi10aUGrW334faoodN7z0ud5JfHr0+e8/ba5OyIiEiyhXBcgIiKZp3AX\nEQkghbuISAAp3EVEAkjhLiISQAp3EZEAUriLiASQwl1EJIAU7iIiAVSQqwNPmTLFZ86cmavDi4jk\npTfeeOOwu5cN1S9n4T5z5kxqampydXgRkbxkZn9Mp58uy4iIBJDCXUQkgBTuIiIBpHAXEQkghbuI\nSAAp3EVEAmjIcDezp83skJm9PcB+M7Mfmtk+M9thZldkvkw5W8/VPsdNv7yJBf+0gJt+eRPP1T6X\n65JEJAvSOXP/KXDzIPtvAeYk/9wD/Pjcy5JMeK72OR559REaWhtwnIbWBh559REFvMh5YMhwd/eX\ngeZBuiwFnvGE14FSM6vIVIFy9p548wnau9pT2tq72nnizSdyVJGIZEsmrrlXAgd6bdcl285gZveY\nWY2Z1TQ1NWXg0DKYxtbGYbWLSHBk9Yaqu69292p3ry4rG3JpBDlH5WPKh9UuIsGRiXCvB2b02p6e\nbJMcu/+K+ykOF6e0FYeLuf+K+3NUkYhkSybCfSOwPDlr5mqgxd0bMvC4co5uq7qNR659hIoxFRhG\nxZgKHrn2EW6rui3XpYnICBtyVUgz+wVwAzDFzOqAbwGFAO6+CngeuBXYB5wC7h6pYmX4bqu6TWEu\nch4aMtzd/a4h9jvwlYxVJCIi50zvUBURCSCFu4hIACncRUQCSOEuIhJACncRkQBSuIuIBJDCXUQk\ngBTuIiIBpHAXEQkghbuISAAp3EVEAkjhLiISQAp3EZEAUriLiASQwl1EJIAU7iIiAaRwFxEJIIW7\niEgAKdxFRAJI4S4iEkAKdxGRAFK4i4gEkMJdRCSAFO4iIgGkcBcRCSCFu4hIACncRUQCSOEuIhJA\nCncRkQBKK9zN7GYz22tm+8zsoX72TzCzTWb2ezPbZWZ3Z75UERFJ15DhbmZh4CngFmA+cJeZze/T\n7SvAO+5+GXAD8A9mVpThWkVEJE3pnLkvBPa5e627dwJrgKV9+jgwzswMGAs0A7GMVioiImlLJ9wr\ngQO9tuuSbb09CVwKHAR2Ave7ezwjFYqIyLBl6obqEuAtYBrwceBJMxvft5OZ3WNmNWZW09TUlKFD\ni4hIX+mEez0wo9f29GRbb3cD6zxhH/AH4JK+D+Tuq9292t2ry8rKzrZmEREZQjrhvg2YY2azkjdJ\nlwEb+/T5AFgMYGYXAvOA2kwWKiIi6SsYqoO7x8zsPmAzEAaedvddZnZvcv8q4DvAT81sJ2DAg+5+\neATrFhGRQQwZ7gDu/jzwfJ+2Vb2+PgjclNnSRETkbOkdqiIiAaRwFxEJIIW7iEgAKdxFRAJI4S4i\nEkAKdxGRAFK4i4gEkMJdRCSAFO4iIgGkcBcRCSCFu4hIACncRUQCSOEuIhJACncRkQBSuIuIBJDC\nXUQkgBTuIiIBpHAXEQkghbuISAAp3EVEAkjhLiISQAp3EZEAUriLiASQwl1EJIAU7iIiAaRwFxEJ\nIIW7iEgAKdxFRAJI4S4iEkAKdxGRAEor3M3sZjPba2b7zOyhAfrcYGZvmdkuM3sps2WKiMhwFAzV\nwczCwFPAp4E6YJuZbXT3d3r1KQV+BNzs7h+Y2dSRKlhERIaWzpn7QmCfu9e6eyewBljap8/ngHXu\n/gGAux/KbJkiIjIc6YR7JXCg13Zdsq23ucBEM/u1mb1hZsszVaCIiAzfkJdlhvE4nwAWAyXAa2b2\nuru/27uTmd0D3ANw0UUXZejQIiLSVzpn7vXAjF7b05NtvdUBm9291d0PAy8Dl/V9IHdf7e7V7l5d\nVlZ2tjWLiMgQ0gn3bcAcM5tlZkXAMmBjnz4bgOvMrMDMLgCuAnZntlQREUnXkJdl3D1mZvcBm4Ew\n8LS77zKze5P7V7n7bjP7d2AHEAd+4u5vj2ThIiIyMHP3nBy4urraa2pqcnJsEZF8ZWZvuHv1UP30\nDlURkQBSuIuIBJDCXUQkgBTuIiIBpHAXEQkghbuISAAp3EVEAkjhLiISQAp3EZEAUriLiASQwl1E\nJIAU7iIiAaRwFxEJIIW7iEgAKdxFRAJI4S4iEkAKdxGRABryY/ZEMmX99noe27yXg8famFZawgNL\n5nH75ZW5LkskkBTukhXrt9ezct1O2qJdANQfa2Plup0ACniREaDLMpIVj23e2xPs3dqiXTy2eW+O\nKhIJNoW7ZMXBY23DaheRc6Nwl6yYVloyrHYROTcKd8mKB5bMo6QwnNJWUhjmgSXzclSRSLDphqpk\nRfdNU82WEckOhbtkze2XVyrMRbJEl2VERAJI4S4iEkAKdxGRAFK4i4gEkMJdRCSA0gp3M7vZzPaa\n2T4ze2iQfleaWczMPpu5EkVEZLiGDHczCwNPAbcA84G7zGz+AP3+Hngh00WKiMjwpHPmvhDY5+61\n7t4JrAGW9tPvq8Ba4FAG6xMRkbOQTrhXAgd6bdcl23qYWSXwZ8CPB3sgM7vHzGrMrKapqWm4tYqI\nSJoydUP1B8CD7h4frJO7r3b3anevLisry9ChRUSkr3SWH6gHZvTanp5s660aWGNmAFOAW80s5u7r\nM1KliIgMSzrhvg2YY2azSIT6MuBzvTu4+6zur83sp8C/KthFRHJnyHB395iZ3QdsBsLA0+6+y8zu\nTe5fNcI1iojIMKW1KqS7Pw8836et31B39y+fe1kiInIu9A5VEZEAUriLiASQwl1EJIAU7iIiAaRw\nFxEJIIW7iEgAKdxFRAJI4S4iEkAKdxGRAFK4i4gEUFrLD8josnvri2xd8wwnjhxm3OQpLFq2nEsX\n3ZjrskRkFFG455ndW1/khdVPEuvsAODE4SZeWP0kgAJeRHroskye2brmmZ5g7xbr7GDrmmdyVJGI\njEYK9zxz4sjhYbWLyPlJ4Z5nxk2eMqx2ETk/KdzzzKJlyykoiqS0FRRFWLRseY4qEpHRSDdU80z3\nTVPNlhGRwSjc89Cli25UmIvIoHRZRkQkgBTuIiIBpHAXEQkghbuISAAp3EVEAkjhLiISQAp3EZEA\nUriLiASQwl1EJIAU7iIiAaTlB+ScNTRuoHb/47R3NFAcqaBq9goqypfmuiyR85rCXc5JQ+MG9ux5\nmHi8DYD2joPs2fMwgAJeJIfSuixjZjeb2V4z22dmD/Wz//NmtsPMdprZq2Z2WeZLldGodv/jPcHe\nLR5vo3b/4zmqSEQgjXA3szDwFHALMB+4y8zm9+n2B+B6d/8Y8B1gdaYLldGpvaNhWO0ikh3pnLkv\nBPa5e627dwJrgJTX2+7+qrsfTW6+DkzPbJkyWhWFLuy3vThSkeVKRKS3dMK9EjjQa7su2TaQvwT+\nrb8dZnaPmdWYWU1TU1P6VcrI2fEsfP+j8Ehp4u8dz6b9T1u3H2LyztuxrqKU9hDFVM1ekelKRWQY\nMnpD1cxuJBHu1/W3391Xk7xkU11d7Zk8tpyFHc/Cpq9BNHnNvOVAYhtgwZ+ndH33t428tmE/J5s7\nGDspwjVLZzPuxQ8Yf+xqvCvO4TlriRUfoaB9MlPrl1HxKd1MFcmldMK9HpjRa3t6si2FmS0AfgLc\n4u5HMlOejKgt3z4d7N2ibYn2XuH+7m8befHne4h1xgE42dzBiz/fw60XhDFgQuO1TGi8NouFi8hQ\n0rkssw2YY2azzKwIWAZs7N3BzC4C1gFfdPd3M1+mjIiWurTaX9uwvyfYu8U647QP8LDh0sgAe0Qk\nW4YMd3ePAfcBm4HdwLPuvsvM7jWze5PdvglMBn5kZm+ZWc2IVSyZM2GA+9592k82d/TbbVdrDCtM\n/RWywhDjl8zMRHUicg7Smufu7s+7+1x3n+3u3022rXL3Vcmv/8rdJ7r7x5N/qkeyaMmQxd+EwpLU\ntsKSRHsvYyf1fybeMq6I0jvm9Jyph0sjlN4xhzGXTx2RckUkfXqH6ijU383LuVeVZ/5A3dfVt3w7\ncSlmwvREsPe5mXrN0tkp19wBCopCXLN0NmMun6owFxmFFO6jzEA3L4GRC/g+Yd5X93Gz8oQjIhmh\ncB9lBrp5+dqG/TkN07lXlSvMRfKIlvwdZQa6eTlQu4hIfxTuo8xANy8Hah9N1jY2U/3qLipefIvq\nV3extrE51yWJnLcU7qPMNUtnU1CU+mPpvnk5mq1tbGbF3gPUdURxoK4jyoq9BxTwIjmicB9l5l5V\nzo2fv6TnTH3spAg3fv6SUX+9+9HaBtriqStKtMWdR2u1OqRILuiG6iiUqZuX67fX89jmvRw81sa0\n0hIeWDKP2y8fbM23s1ffER1Wu4iMLJ25B9T67fWsXLeT+mNtOFB/rI2V63ayfvsZywJlRGWkcFjt\nIjKyFO4B9djmvbRFu1La2qJdPLZ574gcb2VVBSUhS2krCRkrq7Suu0gu6LJMQB081jas9nN1Z/kk\nIHHtvb4jSmWkkJVVFT3tIpJdCveAmlZaQn0/QT6ttKSf3plxZ/kkhbnIKKHLMgH1wJJ5lBSGU9pK\nCsM8sGRejioSkWzSmXtAdc+KydZsGREZXXTmLiISQHl75t7QuIHa/Y/T3tFAcaSCqtkrqCjX53Z2\n654K2T1jpnsqJKCzd5HzQF6Ge0PjBn5W8zD/esw52lXMxHAzf3L0Yb5YjQI+abCpkOmGu55ARfJX\nXl6W+cX277KmGY52hQDjaFeINc2Jdkk416mQDY0b2LPnYdo7DgJOe8dB9ux5mIbGDRmsUkRGSl6G\n+/ojrUQ99Q0zUTfWH2nNUUWjz0BTHtOdClm7/3Hi8dQngni8jdr9j59zbSOloXEDr7yyiC2/+giv\nvLJIT0RyXsvLcE+csafffj4616mQ7R39L/g1UHuu6ZWGSKq8TMOpxROG1X4+uv3ySh6942NUlpZg\nQGVpCY/e8bG0r7cXR/pfNmCg9lzLx1caIiMpL2+o/u2VK/nWK9+gI356xcFIqJC/vXJlDqsafW6/\nvPKsZ8ZUzV7Bnj0PpwRmKFRC1ewVmSovo/LtlYbISMvLcL+t6jYAnnjzCRpbGykfU879V9zf0y7n\nrntWTL7MlimOVCQvyaQKM5HVX7mbE0cOM27yFBYtW86li27MQYUi2WXuPnSvEVBdXe01NTU5OXY+\n2b31RbaueUbhNITua+69X2kYRXzw63KO7B3T01ZQFOGme+7T91Dylpm94e7VQ/XLy2vu54vdW1/k\nhdVPcuJwE7hz4nATL6x+kt1bX8x1aaNORflSLrnkuxRHpgFGcWQajb+dlRLsALHODraueSY3RYpk\nUV5eljlfbF3zDLHOjpS27nDSmeeZKsqXplw2ev1/fQY485XpiSOHs1iVSG7ozH0UGyiEFE7pGTd5\nyrDaRYIkf8/cdzwLW74NLXUwYTos/iYs+POMH6Zl0yYOff8HxBoaKKioYOrf/FcmfOYzGT8OwNrG\n5pQPu/hC6SS6jh45o18uw2nHjh1s2bKFlpYWJkyYwOLFi1mwYEHO6hnMomXLeWH1kymvfgqKIixa\ntjyHVYlkR36G+45nYdPXIJq8edZyILENGQ34lk2baPjGN/H2dgBiBw/S8I1vAmQ84Nc2NrNi7wHa\n4onLCHUdUZ7/xKe4+aUNeLSzp1/vcOr+AGxrXUe08lVOFUcZ11nEl6f/Z+659esZrQ8Swb5p0yai\n0cQU1JaWFv5l/Qb++7t1fOm6q0fdB3V0X7rSDWnJpdbthzi++X26jnUQLo0wfslMxlw+dcSPm9Zs\nGTO7GXgCCAM/cffv9dlvyf23AqeAL7v7m4M95jnNlvn+R9nW/AlesrmctChjvZDr/V2unPQG/M3b\nZ/eY/XjvU4uJHTxzel3BtGnM+dWWjB3nudrneOi1x/HYYeLhybRO+E9c947z1//3GaI4e6ZN4uTk\nqXROnc6M+kYqDxZwYNpNdEQmEfUTbJ29nlNjPuSjRz/KBV0X0BY+RbTzMBNCFzHWjAW79zBjxw4K\nKipouOZqXgoZXaEw4XgX18editdepy5WSe1HltJeWEqko5k5R19m7hVTOPnSyz2vWjb+x8Wc6Ow8\no/4TkRJ+efUSPv3+Lqb/8b20zuj7vkpZWfABd77ywIi/Ehstzhi/PpIwv6R55aB1+yGOrXsPj8Z7\n2qwwROkdc8464NOdLTPkmbuZhYGngE8DdcA2M9vo7u/06nYLMCf55yrgx8m/R8S25k/wZLyamq5K\nWiliDJ3sDI/nrlgRlzVuGNZc7N4rHxaEJ9AVMw6/10Xj78qJlpVQMPkSCiNX42OqiXQ0M7t2I+UN\nb5zxON1n0f19MEbr9kM0Pl9L4YkoHxYb/3uWE/uwleraQqLRnXS1/YZvFFezYNL1lBW9ybiCf6DQ\njnD8kzdyIrqcgpIYr53ayvj9uzgaj3Ns0ljCdohLS97m6rE/56+Oz6fl6F8TZwonaec5+5DVPpZO\nCikYv52Jn95Pxy0hymMHWd78L8w8eAX7S+dTeaCeidu2UTf5cvbMu4t4OAJAR/Fk3pl6K/Et/0z5\nocSTW+zgQU50dIDZGWMf29FGB/Dr8ll84Y/v0dLSwqZNmwD6Dfj+XqWsODURiuZyJwdoKPqQ2roH\naT+8kuLItFE9v/5s9Dv+vQcAFPCj2Okn5E4q2yewMvn7OtiVg+Ob308JdgCPxjm++f0RP3tP54bq\nQmCfu9e6eyewBuj7P20p8IwnvA6UmtmIvU99dXwhv+m6mFYigNFKhN90XczaYx9n1+6/S3s9kb7r\nkUSjxzmyr4u6rRVEWxPrssRCUdqirxHr3ENH8WT2zPsch+YuTnmc7rXT64+14ZxeO3399npatx/i\n8Np3iZyIEgIq2p0H90K4fBLvj91H/OSvubh4Fgun3EJZ0ZtMLHySotARTnVdz8nQVyBSxhutr1PU\n+EfiHgcD95N8pOBn3Dj+KcLMp6XrqzhlGMY4ShjnFcQoomD8W4ypWEt7YTtuRkNhAU+URRgz7TWm\ndh1gwY4dFHR1sb/qT3uCvVs8HGF/1Z+mtF1w6lS/38eTkZKUvwGi0ShbtvT/6ubR2oaeYOvWFi7m\n0ap7ODouzJ6542gvTvxqBnGNmH7HH3cerdW7aUer7ifkuo4ojlFXfCEr5n2dtWXJLIi2Jc7k++g6\n1nFG22DtmZROuFcCB3pt1yXbhtsnY17tqqCL1EWxugjzalcF5u1pryfSdz0SC8Vp+F0ZHuv7bYkR\na/8NkAi9P1SlPrcNtnb68c3vE46l/kcuicO9tZ2MbX4ZiLFg4vUUhAqZUPAMIUv80I/HvoRTDIAf\nfh/z1Gf/68pqKQxFU/p1e5pO4kCkbDPxUGpd7aEQqyaNZam92BPWHZH+zxb7ti/4/Q7CsVhKWzQU\n5rez5gOJM/jeWlpa+n3c+o5o/+2RqZR0OPFw6quDoK0RM+D4B2iX3BvshKRHS90Z/y5cGjmjbbD2\nTMrqVEgzu8fMasyspqmp6awfp5WiQdvTXU+kv37Rk4X9d46fOH2cttRv22Brpw/0DH1huzO2NRF+\nFxSMByBsp6c4dnF6RozFzrzOPb6w44x+3Q4l53Zb4bF+j91YEKbCjnDqggsAiHQ099uvb/vFH3zA\nVe/to3DsOJzEtfaX5n6cfRfOoKArxlW1u1L6T5jQ/0JulZH+v8eVHYeIdMb73RekNWIGHP8A7ZJ7\ng52Q9Jgw/Yz945fMxApT88IKQ4xfMjOT5fUrnXCvB2b02p6ebBtuH9x9tbtXu3t1WVnZcGvtUcqZ\nYQcwKZQI2XRXLuyvX+HYAc6eQuN6vhw7KfVZd7C10wd6hv6w2Dg5JhF+p2LHAejy00Ed5nTQRwpS\n32UJcDwaOaNft6kkznw9WtrvsctjXTT4ZHYsWEAsHGZ27UZCXalPQqGuDmbXbkxps+JirvrC53l4\nxX9jwb1f46Xr/4R9U6dT0NbJdXt/z5ym0z/yLkIsXpx6+arbyqoKSuKpxyvpamdl7Wo6ivr/lRyt\nq1GejZVVFZSEUl+dlISMlVXBGWPQDHZCAkBhSeKmah9jLp9K6R1zenIgXBo5p5upw5FOuG8D5pjZ\nLDMrApYBG/v02Qgst4SrgRZ3H7FTrbLxRygk9XJDETHumP9/cCtOe+XCqtkrCIVOB7PHQ1QsbMIK\n+p49FlBQfF3iq6IQ1yydnbJ3sLXTxy+ZSVdB6n/kthCsqiri5KRPAgXsOPoSsXiUlthy4p74JRhf\n8E8YiSmYV5TeSOK+9mm/aaoiGi9M6dftLygiBHQ0LSEUT/13xfE49zafZEP8BuorL+ToxycwPf4e\nl+z9BcUdR8GdSPsRLtn7z5Qfqum5gVowbRoV3/l2zxTQO8snUXPtf2DVxKmMe62ZD+sLOBkvwh1a\nvYhZV1w/4GyZO8sn8fi4Zqa3f4h5nOntjTy+939yZ9OvaIsYoT7f/tG8GuXZuLN8Eo/Pm8H0SCEG\nTI8U8vi8GbqZOor1+4Qc72Bl7T/ChBnwmR8OOLtrzOVTqXhoIdO/t4iKhxZmJdgh/amQtwI/IDEV\n8ml3/66Z3Qvg7quSUyGfBG4mMRXybncfdJ7juS4c9un/8Y/UnZhCmxcyMXKMO+dsZOFF9Xxs7tcz\nN1umNUxRSSlF4z5JLPoRxk6KcM3S2cy9qvyMxznb2TKx2NvETm3l4uJZXDb5RiYX1jCh6GeEaeKE\n3cbxtuU4Jbx/Yhfbm7fQGW+jozDOW3Mu5JpTxl8UPE8BV9HS9SXiXkZ7qJN1Y7fz8xMXcZwxidky\nUzfSUdBGeayL5UdjzGuv5Mqut/qdwnU2b9oabOyDGmA6mT67VUaj0TJ9Nd2pkFoVUkQkj2hVSBGR\n85jCXUQkgBTuIiIBpHAXEQkghbuISAAp3EVEAkjhLiISQAp3EZEAytmbmMysCfhjBh5qCvSzwEpw\nabzBdj6N93waK2RuvBe7+5CLc+Us3DPFzGrSebdWUGi8wXY+jfd8Gitkf7y6LCMiEkAKdxGRAApC\nuK/OdQFZpvEG2/k03vNprJDl8eb9NXcRETlTEM7cRUSkj7wJdzO72cz2mtk+M3uon/1mZj9M7t9h\nZlfkos5MSWO8n0+Oc6eZvWpml+WizkwYaqy9+l1pZjEz+2w268u0dMZrZjeY2VtmtsvMXsp2jZmU\nxu/yBDPbZGa/T4737lzUmQlm9rSZHTKztwfYn72ccvdR/4fEJ0DtB6qAIuD3wPw+fW4F/g0w4Grg\nt7mue4THey0wMfn1Lfk63nTG2qvfr4Dngc/muu4R/tmWAu8AFyW3p+a67hEe798Bf5/8ugxoBopy\nXftZjveTwBXA2wPsz1pO5cuZ+0Jgn7vXunsnsAbo+7lrS4FnPOF1oNTM8vUTh4ccr7u/6u5Hk5uv\nk/hQ8nyUzs8W4KvAWuBQNosbAemM93PAOnf/AMDd83nM6YzXgXHJj+scSyLcY9ktMzPc/WUS9Q8k\nazmVL+FeCRzotV2XbBtun3wx3LH8JYmzgXw05FjNrBL4M+DHWaxrpKTzs50LTDSzX5vZG2a2PGvV\nZV46430SuBQ4COwE7nf3vp9SHxRZy6mCkXhQyR4zu5FEuF+X61pG0A+AB909nji5C7wC4BPAYqAE\neM3MXnf3d3Nb1ohZArwFfAqYDfw/M9vq7sdzW1Z+y5dwrwdm9Nqenmwbbp98kdZYzGwB8BPgFnc/\nkqXaMi2dsVYDa5LBPgW41cxi7r4+OyVmVDrjrQOOuHsr0GpmLwOXAfkY7umM927ge564KL3PzP4A\nXAL8LjslZlXWcipfLstsA+aY2SwzKwKWARv79NkILE/ejb4aaHH3hmwXmiFDjtfMLgLWAV/M8zO6\nIcfq7rPcfaa7zwR+CfyXPA12SO93eQNwnZkVmNkFwFXA7izXmSnpjPcDEq9SMLMLgXlAbVarzJ6s\n5VRenLm7e8zM7gM2k7j7/rS77zKze5P7V5GYRXErsA84ReJsIC+lOd5vApOBHyXPaGOeh4swpTnW\nwEhnvO6+28z+HdgBxIGfuHu/U+tGuzR/vt8BfmpmO0nMInnQ3fNytUgz+wVwAzDFzOqAbwGFkP2c\n0jtURUQCKF8uy4iIyDAo3EVEAkjhLiISQAp3EZEAUriLiASQwl1EJIAU7iIiAaRwFxEJoP8Pr9zf\n2CAcofIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x99d5ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEKCAYAAAAW8vJGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOBjMzCzDwWBmZhkOBjMzy3AwmJlZxv8HTSMeVFkZJdcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x99d5b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# When to Deploy Feature Scaling - With feature scaling\n",
    "# k_means_cluster.py\n",
    "#!/usr/bin/python \n",
    "\n",
    "\"\"\" \n",
    "    Skeleton code for k-means clustering mini-project.\n",
    "\"\"\"\n",
    "import pickle\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "from feature_format import featureFormat, targetFeatureSplit\n",
    "\n",
    "\n",
    "\n",
    "### load in the dict of dicts containing all the data on each person in the dataset\n",
    "data_dict = pickle.load( open(\"final_project_dataset.pkl\", \"r\") )\n",
    "### there's an outlier--remove it! \n",
    "data_dict.pop(\"TOTAL\", 0)\n",
    "\n",
    "\n",
    "### the input features we want to use \n",
    "### can be any key in the person-level dictionary (salary, director_fees, etc.) \n",
    "feature_1 = \"salary\"\n",
    "feature_2 = \"from_messages\"\n",
    "poi  = \"poi\"\n",
    "features_list = [poi, feature_1, feature_2]\n",
    "data = featureFormat(data_dict, features_list )\n",
    "poi, finance_features = targetFeatureSplit( data )\n",
    "\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "new_finance_features = np.array(finance_features) +0.\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "rescaled_features = scaler.fit_transform(finance_features)\n",
    "\n",
    "\n",
    "### in the \"clustering with 3 features\" part of the mini-project,\n",
    "### you'll want to change this line to \n",
    "### for f1, f2, _ in finance_features:\n",
    "### (as it's currently written, the line below assumes 2 features)\n",
    "for f1, f2 in rescaled_features:\n",
    "    plt.scatter( f1, f2 )\n",
    "plt.show()\n",
    "\n",
    "### cluster here; create predictions of the cluster labels\n",
    "### for the data and store them to a list called pred\n",
    "from sklearn.cluster import KMeans\n",
    "kmeans = KMeans(n_clusters=2, random_state=0)\n",
    "kmeans.fit(rescaled_features)\n",
    "pred = kmeans.predict(rescaled_features)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### rename the \"name\" parameter when you change the number of features\n",
    "### so that the figure gets saved to a different file\n",
    "try:\n",
    "    Draw(pred, rescaled_features, poi, mark_poi=False, name=\"clusters5.jpg\", f1_name=feature_1, f2_name=feature_2)\n",
    "except NameError:\n",
    "    print \"no predictions object named pred found, no clusters to plot\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "86\n",
      "12\n",
      "14368\n"
     ]
    }
   ],
   "source": [
    "from_messages = []\n",
    "\n",
    "for p in data_dict:\n",
    "    if data_dict[p]['from_messages'] == 'NaN':\n",
    "        pass\n",
    "    else:\n",
    "        from_messages.append(data_dict[p]['from_messages'])\n",
    "        \n",
    "    \n",
    "    \n",
    "print len(from_messages)\n",
    "print min(from_messages)\n",
    "print max(from_messages)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}