Python Sets

sets.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tonight we are going to be working with sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'apple', 'bannana', 'grape'}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries = set(['apple','bannana','grape'])\n",
    "groceries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adding more of the same elements yields only the unique values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'apple', 'bannana', 'grape'}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries = set(['apple','bannana','grape','grape','grape'])\n",
    "groceries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can put in any iterable of immutable (unchangable) things.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{' ', 'a', 'g', 'h', 'i', 'n', 'r', 's', 't'}"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_str = set('this is a string')\n",
    "my_str"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Changeable things like lists don't work because lists don't have a __hash__ function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "unhashable type: 'list'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-62-1c684a90f188>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0ma\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'wow'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'i'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'am'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'amazed'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'look'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'at'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'her'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'shoes'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0ma\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mTypeError\u001b[0m: unhashable type: 'list'"
     ]
    }
   ],
   "source": [
    "a = set([['wow','i','am','amazed'],['look','at','her','shoes']])\n",
    "a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "the order isn't kept"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['bannana', 'grape', 'apple']"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries = list(set(['apple','bannana','grape','grape','grape']))\n",
    "groceries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'apple', 'bannana', 'grape'}"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries = set(['apple','bannana','grape'])\n",
    "groceries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "you can ask questions of a set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'orange' in groceries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "last_weeks_groceries = set(['bannana','nuts','pineapple'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "you can compare sets functions like difference, intersection, and union"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'apple', 'grape'}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries.difference(last_weeks_groceries) #what was different this week"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bannana'}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries.intersection(last_weeks_groceries) #what they had in common"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'apple', 'bannana', 'grape', 'nuts', 'pineapple'}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "groceries.union(last_weeks_groceries) #every item from both weeks together"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "But why did Python creators decide to keep both of these strings?  They have the same 4 letters.  They are just in a different order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'asdf', 'fdsa'}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = set(['asdf','fdsa'])\n",
    "b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is an example of how they would change this behaviour.  The eq and hash functions are the ones to pay attention to in a class that works with sets.  Here is our normal example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "###WARNING:  don't mess with Python internals\n",
    "###   this is a short example of how Python Dev's did it.\n",
    "class My_String():\n",
    "    def __init__(self, s):\n",
    "        self.s = s\n",
    "\n",
    "    def __eq__(self, other):\n",
    "        return self.s == other.s  #in our next case 'asdf' doesn't equal 'fdsa' so this returns False\n",
    "    \n",
    "    def __hash__(self):\n",
    "        return hash(self.s)\n",
    "\n",
    "    def __repr__(self): #not needed for this example... just easier to read the answers\n",
    "        return self.s\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "which gives us the normal result keeping both strings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{asdf, fdsa}"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = set([My_String('asdf'),My_String('fdsa')])\n",
    "b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### But If you wanted to change that behaviour  we could change these functions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The one important thing to note is:\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## __hash__ and __eq__ need to have a special relationship\n",
    "## the rule is, if x == y, then hash(x) == hash(y)\n",
    "## if you don't make that rule work, then things act weird.\n",
    "\n",
    "#### to make it easier make sure to use the same variable in __hash__ and __eq__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class My_String():\n",
    "    def __init__(self, s):\n",
    "        self.s = s\n",
    "\n",
    "    def __eq__(self, other):\n",
    "        return set(self.s)==set(other.s) #in our case {'a','s','d','f'}=={'s','a','f','d'}\n",
    "    \n",
    "    def __hash__(self):\n",
    "        return hash(tuple(sorted(set(self.s)))) #but this gets complicated\n",
    "        \n",
    "    def __repr__(self): #not needed for this example... just easier to read the answers\n",
    "        return self.s\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I needed to turn it into something hashable which is why I turned the set of letters into a sorted tuple.\n",
    "\n",
    "But now My_String class says \"if the string is made of the same letters than don't put a second instance in a set\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{asdf, qwerty}"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = set([My_String('asdf'), My_String('fdsa'), My_String('asdfasdfasfdfasdf'), My_String('qwerty')])\n",
    "b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "another example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class Alien():\n",
    "    def __init__(self, arms, name):\n",
    "        self.arms = arms\n",
    "        self.name = name\n",
    "\n",
    "    def __eq__(self, other):\n",
    "        return self.arms == other.arms\n",
    "    \n",
    "    def __hash__(self):\n",
    "        return hash(self.arms) #make sure you are hashing the thing you are comparing in __eq__\n",
    "\n",
    "    def __repr__(self): #not needed for this example... just easier to read the answers\n",
    "        return self.name\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you have seen a one arm alien you've seen them all."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{klaatu, Ewok}"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c = set([Alien(1, 'klaatu'), Alien(1, 'barada'), Alien(1, 'nikto'), Alien(2, 'Ewok'),  Alien(2, 'Wookie'), ])\n",
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}