PythagorasMatrix.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pythagoras via Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import math\n",
    "import random\n",
    "import operator\n",
    "import warnings\n",
    "import numpy as np\n",
    "\n",
    "from geneticprogramming import Baseset\n",
    "from geneticprogramming import Individual\n",
    "from geneticprogramming import Population\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training data \n",
    "\n",
    "* a number of random matrices (2D floating points) on a plane"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "SAMPLE_SIZE = 50\n",
    "PLANE_SIZE = 20.0\n",
    "RANDOMPOINTS = [PLANE_SIZE * np.random.random_sample((1, 2)) for _ in range(SAMPLE_SIZE)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Base functions\n",
    "\n",
    "* _get_ a value at a location from a matrix\n",
    "* an _ephemeral integer_ - either 0 or 1\n",
    "* _addition_ and _subtraction_ of floats\n",
    "* _square_ and _square root_ of floats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def getValue(ndarray, idx):\n",
    "    \"\"\"Return the indexed value from the 1x2 numpy array\"\"\"\n",
    "    return ndarray[0][idx]\n",
    "\n",
    "square = lambda x: x ** 2\n",
    "sqrt = lambda x: math.sqrt(abs(x))\n",
    "\n",
    "bset = Baseset()\n",
    "bset.add_ephemeral('E', lambda: random.randint(0, 1), int)\n",
    "bset.add_primitive(getValue, [np.ndarray, int], float, name=\"get\")\n",
    "bset.add_primitive(operator.add, [float, float], float, name=\"add\")\n",
    "bset.add_primitive(operator.sub, [float, float], float, name=\"sub\")\n",
    "bset.add_primitive(square, [float], float, name=\"square\")\n",
    "bset.add_primitive(sqrt, [float], float, name=\"sqrt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Individuals\n",
    "* consume a matrix (a 1x2 array of floats - x and y coordinates)\n",
    "* return a float (the distance from the origin)\n",
    "* scored by accumulating the error when applying the Training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Individual.INTYPES = [np.ndarray]\n",
    "Individual.OUTTYPE = float\n",
    "\n",
    "def evaluate(individual):\n",
    "    \"\"\"sum of application of all the random points\"\"\"\n",
    "    program = individual.compile()\n",
    "    score = 0\n",
    "    try:\n",
    "        for point in RANDOMPOINTS:\n",
    "            program_distance = program(point)\n",
    "            true_distance = math.hypot(point[0][0], point[0][1])\n",
    "            score += abs(true_distance - program_distance)\n",
    "    except (OverflowError, RuntimeWarning):\n",
    "        pass\n",
    "    if math.isnan(score) or score == 0:\n",
    "        score = 100000\n",
    "\n",
    "    # accumulate the number of nodes actually used during a run by calling the adfs in the rpb\n",
    "    nodes = 0\n",
    "    for node in individual.trees[-1]:\n",
    "        if node.name[:1] != 'F':\n",
    "            nodes += 1\n",
    "        else:\n",
    "            nodes += len(individual.trees[int(node.name[1])])\n",
    "    modifier = 1 + (-2 ** - (nodes / 250))\n",
    "\n",
    "    return score + modifier,\n",
    "Individual.evaluate = evaluate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Population Evolution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gen\tavg\tstd\tmin    \tmax\n",
      "0  \tinf\tnan\t169.498\tinf\n",
      "1  \t5.62833e+08\t7.93964e+09\t15.8184\t1.12565e+11\n",
      "2  \t423.53     \t916.245    \t15.8184\t9910.72    \n",
      "3  \t883.638    \t2153.08    \t0.0273451\t11996.7    \n",
      "4  \t990.413    \t2509.91    \t0.0273451\t11996.7    \n",
      "5  \t990.378    \t2656.07    \t0.0273451\t11983.1    \n",
      "6  \t994.654    \t2523.49    \t0.0273451\t11507.4    \n",
      "7  \t998.007    \t2384.86    \t0.0273451\t11507.4    \n",
      "8  \t1015.93    \t2446.26    \t0.0273451\t11507.4    \n",
      "9  \t1306.51    \t2900.85    \t0.0273451\t11507.4    \n",
      "10 \t1192.83    \t2905.12    \t0.0273451\t11507.4    \n",
      "11 \t1010.79    \t2492.82    \t0.0273451\t11507.4    \n",
      "12 \t963.233    \t2407.06    \t0.0273451\t11507.4    \n",
      "13 \t1116.6     \t2693.88    \t0.0273451\t11507.4    \n",
      "14 \t1346.5     \t2965.82    \t0.0273451\t11507.4    \n"
     ]
    }
   ],
   "source": [
    "Population.POPULATION_SIZE = 200       # Number of individuals in a generation\n",
    "Population.MATE_MUTATE_CLONE = (60, 38, 2)   # ratio of individuals to mate, mutate, or clone\n",
    "Population.CLONE_BEST = 10             # Number of best individuals to seed directly into offspring\n",
    "\n",
    "Individual.MAX_ADFS = 4                # The maximum number of ADFs to generate\n",
    "Individual.ADF_NARGS = (1, 5)          # min, max number of input arguments to adfs\n",
    "Individual.GROWTH_TERM_PB = 0.3        # Probability of terminal when growing:\n",
    "Individual.GROWTH_MAX_INIT_DEPTH = 5   # Maximum depth of initial growth\n",
    "Individual.GROWTH_MAX_MUT_DEPTH = 3    # Maximum depth of mutation growth\n",
    "\n",
    "NUM_GENERATIONS = 15\n",
    "\n",
    "population = Population(bset)\n",
    "for generation in range(NUM_GENERATIONS):\n",
    "    population.evolve()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABmsAAAGRCAYAAACKdIeEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYxTFhAyYLAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f198cdb6400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best = population[0]\n",
    "best.draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.4.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}