generic gradient descent 2-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17.801836954415\n",
      "14.13767474703819\n",
      "11.29899783056143\n",
      "9.099086589002757\n",
      "7.393468961548655\n",
      "6.070361209639232\n",
      "5.043271627886479\n",
      "4.245277794847336\n",
      "3.6245986959077263\n",
      "3.1411687307532485\n",
      "2.7639869114691984\n",
      "2.4690658507965226\n",
      "2.237844827504499\n",
      "2.0559619232961928\n",
      "1.9123039850184107\n",
      "1.7982715493318602\n",
      "1.7072100908183623\n",
      "1.633969959934321\n",
      "1.5745658924793733\n",
      "1.5259135607824263\n",
      "1.485625734569154\n",
      "1.4518545637788032\n",
      "1.4231695474341495\n",
      "1.3984631139729506\n",
      "1.3768775654671266\n",
      "1.357748551779016\n",
      "1.3405613344701508\n",
      "1.3249169465602708\n",
      "1.3105060090303895\n",
      "1.2970884716006865\n",
      "1.2844779373156532\n",
      "1.272529533772799\n",
      "1.2611305285081957\n",
      "1.2501930676281967\n",
      "1.2396485572690117\n",
      "1.2294433161685026\n",
      "1.2195352117419418\n",
      "1.2098910571296522\n",
      "1.2004845970362976\n",
      "1.1912949491403273\n",
      "1.1823053979955995\n",
      "1.1735024616705918\n",
      "1.164875169416247\n",
      "1.156414502616237\n",
      "1.1481129620768251\n",
      "1.1399642330724125\n",
      "1.1319629260304402\n",
      "1.1241043757434257\n",
      "1.1163844858678047\n",
      "1.108799608465264\n",
      "1.1013464506597834\n",
      "1.0940220022775444\n",
      "1.086823479724182\n",
      "1.079748282427757\n",
      "1.072793959006461\n",
      "1.0659581809628855\n",
      "1.059238722204041\n",
      "1.052633443071037\n",
      "1.0461402778602253\n",
      "1.0397572250477922\n",
      "1.0334823396081974\n",
      "1.027313726954603\n",
      "1.021249538136424\n",
      "1.0152879660112697\n",
      "1.0094272421728923\n",
      "1.003665634465861\n",
      "0.9980014449560416\n",
      "0.992433008255606\n",
      "0.9869586901240717\n",
      "0.9815768862847122\n",
      "0.9762860214092679\n",
      "0.9710845482346172\n",
      "0.965970946783183\n",
      "0.9609437236652277\n",
      "0.9560014114461161\n",
      "0.9511425680654627\n",
      "0.9463657762978712\n",
      "0.9416696432474893\n",
      "0.9370527998701739\n",
      "0.9325139005184827\n",
      "0.9280516225057897\n",
      "0.9236646656866032\n",
      "0.9193517520508592\n",
      "0.9151116253303571\n",
      "0.910943050615955\n",
      "0.9068448139844297\n",
      "0.902815722134103\n",
      "0.8988546020285535\n",
      "0.894960300547833\n",
      "0.8911316841467161\n",
      "0.8873676385196635\n",
      "0.8836670682720946\n",
      "0.8800288965978147\n",
      "0.8764520649622604\n",
      "0.8729355327915019\n",
      "0.8694782771667173\n",
      "0.866079292524036\n",
      "0.8627375903596515\n",
      "0.8594521989400123\n",
      "0.8562221630170561\n",
      "0.853046543548309\n",
      "0.8499244174217807\n",
      "0.8468548771856245\n",
      "0.8438370307823253\n",
      "0.8408700012875013\n",
      "0.837952926653117\n",
      "0.8350849594550573\n",
      "0.8322652666450348\n",
      "0.8294930293066556\n",
      "0.8267674424157024\n",
      "0.8240877146044279\n",
      "0.8214530679298782\n",
      "0.8188627376461405\n",
      "0.8163159719804728\n",
      "0.8138120319132054\n",
      "0.8113501909614202\n",
      "0.8089297349662457\n",
      "0.8065499618837952\n",
      "0.8042101815796665\n",
      "0.8019097156268893\n",
      "0.7996478971073286\n",
      "0.7974240704164391\n",
      "0.7952375910713514\n",
      "0.7930878255222101\n",
      "0.7909741509666827\n",
      "0.7888959551676629\n",
      "0.7868526362740256\n",
      "0.7848436026444526\n",
      "0.7828682726742306\n",
      "0.7809260746249893\n",
      "0.7790164464573134\n",
      "0.7771388356662474\n",
      "0.7752926991195194\n",
      "0.7734775028985575\n",
      "0.7716927221421871\n",
      "0.7699378408929883\n",
      "0.7682123519462466\n",
      "0.7665157567014623\n",
      "0.7648475650164078\n",
      "0.763207295063645\n",
      "0.7615944731894425\n",
      "0.760008633775153\n",
      "0.7584493191008871\n",
      "0.7569160792115318\n",
      "0.7554084717850068\n",
      "0.7539260620028032\n",
      "0.7524684224226822\n",
      "0.7510351328535461\n",
      "0.749625780232431\n",
      "0.7482399585035865\n",
      "0.7468772684995979\n",
      "0.7455373178245411\n",
      "0.7442197207391179\n",
      "0.7429240980477125\n",
      "0.7416500769874018\n",
      "0.7403972911188403\n",
      "0.7391653802189567\n",
      "0.7379539901755295\n",
      "0.7367627728835066\n",
      "0.7355913861431133\n",
      "0.7344394935596634\n",
      "0.7333067644450891\n",
      "0.7321928737211386\n",
      "0.7310975018241933\n",
      "0.7300203346117362\n",
      "0.7289610632703892\n",
      "0.7279193842254896\n",
      "0.7268949990522503\n",
      "0.7258876143883659\n",
      "0.7248969418481672\n",
      "0.7239226979381617\n",
      "0.722964603974078\n",
      "0.7220223859992493\n",
      "0.721095774704446\n",
      "0.7201845053490126\n",
      "0.7192883176833821\n",
      "0.7184069558728846\n",
      "0.7175401684228391\n",
      "0.71668770810496\n",
      "0.7158493318849593\n",
      "0.7150248008513975\n",
      "0.7142138801457741\n",
      "0.7134163388937281\n",
      "0.712631950137488\n",
      "0.71186049076942\n",
      "0.7111017414667101\n",
      "0.7103554866271574\n",
      "0.7096215143060748\n",
      "0.7088996161542158\n",
      "0.7081895873568059\n",
      "0.7074912265735727\n",
      "0.7068043358797983\n",
      "0.7061287207084019\n",
      "0.7054641897929553\n",
      "0.7048105551117031\n",
      "0.7041676318325125\n",
      "0.7035352382587838\n",
      "0.7029131957762347\n",
      "0.7023013288006239\n",
      "0.7016994647263548\n",
      "0.7011074338759381\n",
      "0.7005250694503242\n",
      "0.6999522074800654\n",
      "0.6993886867773232\n",
      "0.6988343488886781\n",
      "0.6982890380487481\n",
      "0.6977526011345826\n",
      "0.6972248876208327\n",
      "0.696705749535691\n",
      "0.6961950414175846\n",
      "0.6956926202725742\n",
      "0.6951983455324912\n",
      "0.6947120790138129\n",
      "0.6942336848771755\n",
      "0.6937630295876358\n",
      "0.6932999818755647\n",
      "0.6928444126982206\n",
      "0.6923961952019784\n",
      "0.6919552046851808\n",
      "0.6915213185616492\n",
      "0.6910944163247847\n",
      "0.6906743795122878\n",
      "0.6902610916714975\n",
      "0.6898544383252756\n",
      "0.6894543069385193\n",
      "0.689060586885189\n",
      "0.6886731694159428\n",
      "0.6882919476262966\n",
      "0.6879168164253294\n",
      "0.6875476725049006\n",
      "0.6871844143094407\n",
      "0.6868269420061852\n",
      "0.6864751574559782\n",
      "0.6861289641845167\n",
      "0.685788267354132\n",
      "0.6854529737360073\n",
      "0.68512299168289\n",
      "0.6847982311022837\n",
      "0.6844786034300221\n",
      "0.684164021604407\n",
      "0.6838544000406627\n",
      "0.6835496546059222\n",
      "0.6832497025945804\n",
      "0.6829544627040809\n",
      "0.6826638550111092\n",
      "0.6823778009482194\n",
      "0.6820962232808107\n",
      "0.6818190460845195\n",
      "0.6815461947230045\n",
      "0.6812775958260862\n",
      "0.6810131772682634\n",
      "0.6807528681476179\n",
      "0.6804965987650453\n",
      "0.680244300603841\n",
      "0.6799959063096528\n",
      "0.6797513496707785\n",
      "0.6795105655987473\n",
      "0.6792734901092913\n",
      "0.6790400603036041\n",
      "0.6788102143499289\n",
      "0.6785838914654628\n",
      "0.678361031898548\n",
      "0.6781415769111996\n",
      "0.6779254687618878\n",
      "0.6777126506886539\n",
      "0.6775030668924763\n",
      "0.6772966625209451\n",
      "0.6770933836522003\n",
      "0.6768931772791554\n",
      "0.676695991293964\n",
      "0.6765017744727737\n",
      "0.6763104764607272\n",
      "0.6761220477572145\n",
      "0.6759364397014118\n",
      "0.6757536044579859\n",
      "0.6755734950031329\n",
      "0.6753960651108006\n",
      "0.6752212693391347\n",
      "0.675049063017203\n",
      "0.6748794022318966\n",
      "0.6747122438150984\n",
      "0.6745475453310205\n",
      "0.6743852650637877\n",
      "0.6742253620052375\n",
      "0.6740677958429006\n",
      "0.6739125269482078\n",
      "0.6737595163648767\n",
      "0.6736087257975194\n",
      "0.6734601176004237\n",
      "0.6733136547665424\n",
      "0.6731693009166405\n",
      "0.6730270202886595\n",
      "0.6728867777272519\n",
      "0.6727485386734723\n",
      "0.6726122691546677\n",
      "0.6724779357745427\n",
      "0.6723455057033544\n",
      "0.6722149466683388\n",
      "0.6720862269442299\n",
      "0.6719593153439909\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.datasets import load_boston\n",
    "import matplotlib.pyplot as plt\n",
    "data=load_boston()\n",
    "\n",
    "def step_gradient(X, Y, m, learning_rate):\n",
    "    m_slope=np.zeros(len(X[0]))\n",
    "    for i in range(len(X)):\n",
    "        x=X[i]\n",
    "        y=Y[i]\n",
    "        for j in range(len(x)):\n",
    "            m_slope[j]+=(-2/len(x))*(y-sum(m*x))*x[j]\n",
    "    new_m=m-(learning_rate*m_slope)\n",
    "    return new_m\n",
    "\n",
    "def cost(m, x, y):\n",
    "    cost=0\n",
    "    for i in range(len(x)):\n",
    "        cost+=(1/len(x))*(y[i]-sum(m*x[i]))\n",
    "    print(cost)\n",
    "    \n",
    "\n",
    "def gd(x, y, learning_rate, iterations):\n",
    "    m=np.zeros(len(x[0]))\n",
    "    for i in range(iterations):\n",
    "        m=step_gradient(x, y, m, learning_rate)\n",
    "        cost(m, x, y)\n",
    "    return m\n",
    "\n",
    "def run():\n",
    "    iterations=300\n",
    "    learning_rate=0.00000001\n",
    "    x=np.append(data.data, np.ones(len(data.data)).reshape(-1, 1), axis=1)\n",
    "    y=data.target\n",
    "    m=gd(x, y, learning_rate, iterations)\n",
    "run()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'm' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-2-9a40b379906c>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mm\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'm' is not defined"
     ]
    }
   ],
   "source": [
    "m"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}