UpperConfidenceBound.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('Ads_CTR_Optimisation.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "N = 1000\n",
    "d = 10\n",
    "ads_selected = []\n",
    "no_of_selections = [0]*d\n",
    "sum_of_rewards = [0]*d\n",
    "total_reward = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "for n in range (0,N):\n",
    "    ad = 0\n",
    "    max_upper_bound = 0\n",
    "    for i in range(0,d):\n",
    "        if(no_of_selections[i]>0):\n",
    "            avg_reward = sum_of_rewards[i]/no_of_selections[i]\n",
    "            delta_i = math.sqrt(3/2 * math.log(n+1)/no_of_selections[i])\n",
    "            upper_bound = avg_reward+delta_i\n",
    "        else:\n",
    "            upper_bound = 1e400\n",
    "        if upper_bound > max_upper_bound:\n",
    "            max_upper_bound = upper_bound\n",
    "            ad = i\n",
    "    ads_selected.append(ad)\n",
    "    no_of_selections[ad] = no_of_selections[ad]+1\n",
    "    reward = df.values[n,ad]\n",
    "    sum_of_rewards[ad] = sum_of_rewards[ad]+reward\n",
    "    total_reward = total_reward+reward"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(ads_selected)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}