master-thesis/EvolutionaryAlgorithm/ea_tools.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Statistical Tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import scipy.stats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Helper function to calculate the wheel of fortune"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wheel_of_fortune(rank_i,n):\n",
    "    return rank_i / (0.5 * n * (n + 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wheel_of_fortune_weights(items:list, item_scores:list):\n",
    "    rank = scipy.stats.rankdata(item_scores)\n",
    "\n",
    "    n = len(items)\n",
    "\n",
    "    return wheel_of_fortune(rank, n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wheel_of_fortune_selection(items: list, item_scores:list, num_choices=1):\n",
    "    \n",
    "    wheel_weights = wheel_of_fortune_weights(items, item_scores)\n",
    "    \n",
    "    n = min(len(items), num_choices)\n",
    "    \n",
    "    choice = np.random.choice(items, size=n, replace=False, p=wheel_weights)\n",
    "    \n",
    "    if num_choices == 1:\n",
    "        return choice[0]\n",
    "\n",
    "    return choice\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def combined_wheel_of_fortune_selection(items_list:list, item_scores_list:list, num_choices=1):\n",
    "    \n",
    "    scores = {}\n",
    "    \n",
    "    for i in range(len(items_list)):\n",
    "        items = items_list[i]\n",
    "        item_scores = item_scores_list[i]\n",
    "        \n",
    "        w = wheel_of_fortune_weights(items, item_scores)\n",
    "        #print(items, item_scores)\n",
    "        #print(w)\n",
    "        \n",
    "        for j, item in enumerate(items):\n",
    "            if item in scores:\n",
    "                scores[item] += w[j]\n",
    "            else:\n",
    "                scores[item] = w[j]\n",
    "        \n",
    "    combined_items = []\n",
    "    combined_scores = []\n",
    "    \n",
    "    for i,s in scores.items():\n",
    "        combined_items.append(i)\n",
    "        combined_scores.append(s)\n",
    "    \n",
    "    combined_scores = np.array(combined_scores)\n",
    "    \n",
    "    #print(combined_scores)\n",
    "    #print(np.sum(combined_scores))\n",
    "    \n",
    "    combined_scores /= len(items_list)\n",
    "    \n",
    "    #print(combined_scores)\n",
    "    \n",
    "    #print(np.sum(combined_scores))\n",
    "    \n",
    "    n = min(len(combined_items), num_choices)\n",
    "    \n",
    "    return np.random.choice(combined_items, size=n, replace=False, p=combined_scores)\n",
    "        \n",
    "        "
   ]
  }
 ],
 "metadata": {
  "file_extension": ".py",
  "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": "3.7.5"
  },
  "mimetype": "text/x-python",
  "name": "python",
  "npconvert_exporter": "python",
  "pygments_lexer": "ipython3",
  "version": 3
 },
 "nbformat": 4,
 "nbformat_minor": 4
}