testingGP.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/jupyter/anaconda3/lib/python3.6/site-packages/deap/creator.py:141: RuntimeWarning: A class named 'FitnessMax' has already been created and it will be overwritten. Consider deleting previous creation of that class or rename it.\n",
      "  RuntimeWarning)\n",
      "/home/jupyter/anaconda3/lib/python3.6/site-packages/deap/creator.py:141: RuntimeWarning: A class named 'Individual' has already been created and it will be overwritten. Consider deleting previous creation of that class or rename it.\n",
      "  RuntimeWarning)\n"
     ]
    }
   ],
   "source": [
    "import operator\n",
    "import math\n",
    "import game\n",
    "import marshal\n",
    "import random\n",
    "import time\n",
    "from deap import base\n",
    "from deap import creator\n",
    "from deap import gp\n",
    "from deap import tools\n",
    "from deap import algorithms\n",
    "\n",
    "from prettyPrintTree import prettyPrint\n",
    "\n",
    "from contextlib import redirect_stdout\n",
    "\n",
    "import multiprocessing as mp\n",
    "\n",
    "#toolbox.evaluate \n",
    "def evalFunc(individual, n):\n",
    "\t#print (prettyPrint(individual))\n",
    "\t#print (individual)\n",
    "\t#UCBFunctionToGet = toolbox.compile(individual)\n",
    "\t#pass UCBFunctionToGet\n",
    "\t#play game n times with n seeds, return avg score overall\n",
    "\t#n = 1\n",
    "\tscore = game.main([0,n], individual, False)\n",
    "\n",
    "\treturn score,\n",
    "\n",
    "def originalMCTSFunc():\n",
    "\tnew_individual = gp.PrimitiveTree.from_string(\"add(truediv(child_win_score, child_visit_count), (mul(1.414,sqrt(mul(2.0,truediv(log(current_visit_count),child_visit_count))))) )\", pset)\n",
    "\n",
    "\t#UCBFunctionToGet = toolbox.compile(new_individual)\n",
    "\t#print(UCBFunctionToGet(2,3,4),)\n",
    "\n",
    "\treturn new_individual\n",
    "\n",
    "pool = mp.Pool()\n",
    "pset = gp.PrimitiveSetTyped('MAIN', [float, float, float], float)\n",
    "pset.renameArguments(ARG0='child_win_score', ARG1='child_visit_count', ARG2='current_visit_count')\n",
    "pset.addPrimitive(operator.add, [float, float], float)\n",
    "pset.addPrimitive(operator.mul, [float, float], float)\n",
    "pset.addPrimitive(operator.truediv, [float, float], float)\n",
    "pset.addPrimitive(math.sqrt, [float], float)\n",
    "pset.addPrimitive(math.log, [float],float)\n",
    "pset.addTerminal(1.414, float)\n",
    "pset.addTerminal(0.5, float)\n",
    "pset.addTerminal(2.0, float)\n",
    "pset.addTerminal(3.0, float)\n",
    "pset.addTerminal(4.0, float)\n",
    "pset.addTerminal(5.0, float)\n",
    "\n",
    "creator.create('FitnessMax', base.Fitness, weights=(1.0,))\n",
    "creator.create('Individual', gp.PrimitiveTree, fitness=creator.FitnessMax,\n",
    "              pset=pset)\n",
    "\n",
    "toolbox = base.Toolbox()\n",
    "toolbox.register('compile', gp.compile, pset=pset)\n",
    "toolbox.register('expr', gp.genFull, pset=pset, min_=5, max_=5)\n",
    "toolbox.register('individual', tools.initIterate, creator.Individual,\n",
    "                toolbox.expr)\n",
    "\n",
    "num_sims = 200\n",
    "toolbox.register('evaluate', evalFunc, n=num_sims)\n",
    "\n",
    "toolbox.register('population', tools.initRepeat, list, toolbox.individual)\n",
    "toolbox.register('select', tools.selBest)\n",
    "toolbox.register('mate', gp.cxOnePoint)\n",
    "toolbox.register('mutate', gp.mutUniform, expr=toolbox.expr, pset=pset)\n",
    "\n",
    "toolbox.register('map', pool.map)\n",
    "\n",
    "#for the single uct individual\n",
    "toolbox.register('create_initial_uct', originalMCTSFunc)\n",
    "toolbox.register('initial_uct', tools.initIterate, creator.Individual, toolbox.create_initial_uct)\n",
    "\n",
    "initial_uct_indiv = toolbox.initial_uct()\n",
    "#s =toolbox.evaluate(initial_uct_indiv)\n",
    "#print(s)\n",
    "\n",
    "#then test generations\n",
    "#population of 5 so computer doesnt cry\n",
    "pop_size = 99\n",
    "pop = toolbox.population (pop_size-1)\n",
    "pop.append(initial_uct_indiv)\n",
    "\n",
    "#final_pop = algorithms.eaSimple(pop, toolbox=toolbox, cxpb=0.5, mutpb=0.2, ngen=2)\n",
    "\n",
    "cxpb=0.5\n",
    "mutpb=0.2\n",
    "ngen=100\n",
    "\n",
    "func_globals = globals()\n",
    "func_globals['add'] = operator.add\n",
    "func_globals['mul'] = operator.mul\n",
    "func_globals['truediv'] = operator.truediv\n",
    "func_globals['sqrt'] = math.sqrt\n",
    "func_globals['log'] = math.log\n",
    "\n",
    "compiled_pop = []\n",
    "for i in range(len(pop)):\n",
    "    UCBFunc_code = marshal.dumps(toolbox.compile(pop[i]).__code__)\n",
    "\n",
    "    compiled_pop.append(UCBFunc_code)\n",
    "\n",
    "output_file = open('output.csv.txt', 'w')\n",
    "output_file.write('GEN;num-sims;pop-size;max-fitness;ellapsed-time;\\n')\n",
    "output_file.close()\n",
    "\n",
    "evals = toolbox.map(toolbox.evaluate, compiled_pop)\n",
    "for i in range(len(compiled_pop)):\n",
    "    pop[i].fitness = evals[i]\n",
    "\n",
    "for g in range(ngen):\n",
    "    current_time = time.time()\n",
    "    pop = toolbox.select(pop, len(pop))\n",
    "    \n",
    "    offspring = toolbox.clone(pop)\n",
    "    i = 0\n",
    "    while i < len(offspring) - 1:\n",
    "        random_n = random.uniform(0, 1)\n",
    "        if random_n <= cxpb:\n",
    "            ind1, ind2 = toolbox.mate(offspring[i], offspring[i+1])\n",
    "            offspring[i] = ind1\n",
    "            offspring[i+1] = ind2\n",
    "        \n",
    "        i += 2\n",
    "    \n",
    "    for i in range(len(offspring)):\n",
    "        random_n = random.uniform(0, 1)\n",
    "        if random_n <= mutpb:\n",
    "            offspring[i] = toolbox.mutate(offspring[i])[0]\n",
    "    \n",
    "    compiled_offspring = []\n",
    "    for i in range(len(offspring)):\n",
    "        UCBFunc_code = marshal.dumps(toolbox.compile(offspring[i]).__code__)\n",
    "\n",
    "        compiled_offspring.append(UCBFunc_code)\n",
    "        \n",
    "    evals = toolbox.map(toolbox.evaluate, compiled_offspring)\n",
    "    for i in range(len(compiled_offspring)):\n",
    "        offspring[i].fitness = evals[i]\n",
    "        \n",
    "    pop = offspring\n",
    "    \n",
    "    output_file = open('output.csv.txt', 'a')\n",
    "    output_file.write(str(g) + ';' + str(num_sims) + \";\" + str(pop_size) + \";\" + str(max([p.fitness for p in pop])) + ';' + str(time.time()-current_time) + ';\\n')\n",
    "    output_file.close()\n",
    "    \n",
    "    results_file = open('gen-results.csv.txt', 'w')\n",
    "    results_file.write('Individual;Fitness;')\n",
    "    for fp in toolbox.select(pop, k = 3):\n",
    "        results_file.write(str(fp) + \";\" + str(fp.fitness) + \";\\n\")\n",
    "    results_file.close()\n",
    "\n",
    "pool.terminate()\n"
   ]
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}