Ticket Comparison Solver.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# before we do the work, order the tickets in order of size\n",
    "# each row is a team member producing a sequence prediction\n",
    "\n",
    "sequences_of_ticket_ids = [\n",
    "    [1, 2, 3, 4, 5, 6, 7],\n",
    "    [1, 4, 3, 6, 5, 7, 2],\n",
    "    [2, 3, 6, 7, 4, 5, 1],\n",
    "    [3, 5, 7],\n",
    "    [4, 7, 3],\n",
    "]\n",
    "\n",
    "# the team did some work and the hours were recorded against tickets\n",
    "\n",
    "recorded_hours = {\n",
    "    '1': 5,\n",
    "    '2': 7,\n",
    "    '4': 13,\n",
    "    '3': 20,\n",
    "    '5': 30,\n",
    "    '6': 24,\n",
    "    '7': 32\n",
    "}\n",
    "\n",
    "# now we've done some work, we've introduced Ticket 8 and some more sequences based on size comparisons of past work\n",
    "\n",
    "new_sequences_of_ticket_ids = [\n",
    "    [2, 8, 7],\n",
    "    [1, 8, 6],\n",
    "    [3, 8, 5],\n",
    "    [8, 4, 5],\n",
    "    [8, 6, 5]\n",
    "]\n",
    "\n",
    "sequences_lists = [sequences_of_ticket_ids, new_sequences_of_ticket_ids]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write a function thingy that predicts the hourly value of Ticket 8?\n",
    "\n",
    "def prediction_thingy(id, sequences_lists, recorded_hours):\n",
    "    \n",
    "    hours = None\n",
    "\n",
    "    # range boundaries are zero for the moment; but we could expect these to \n",
    "    # be influenced by the recorded hours\n",
    "    upper_limit = 0\n",
    "    upper_limits = []\n",
    "    \n",
    "    lower_limit = 0\n",
    "    lower_limits = []\n",
    "    \n",
    "    \n",
    "    for sequence_list in sequences_lists:\n",
    "        \n",
    "        for sequence in sequence_list:\n",
    "            \n",
    "            if id in sequence:\n",
    "                # at this point we're only interested in the sequences that contain\n",
    "                # the ticket id were seeking a prediction on.\n",
    "                \n",
    "                # Our resolution strategy is to take an average of the upper and lower limits\n",
    "                # defined by the hours recorded on comparison tickets; then we'll split the \n",
    "                # middle to derrive an hourly estimate for the ticket.\n",
    "                \n",
    "                \n",
    "                # retrieve the first index of the ticket id in the list\n",
    "                id_index = sequence.index(id)\n",
    "                \n",
    "                # find the nearest ticket that is predicted to be smaller than the ticket\n",
    "                if id_index > 0:\n",
    "                    lower_id = sequence[id_index - 1]\n",
    "                    lower_hours = recorded_hours[str(lower_id)]\n",
    "                    lower_limits.append(lower_hours)\n",
    "                    \n",
    "                # find the nearest ticket that is predicted to be higher than the ticket\n",
    "                if (id_index + 1) < len(sequence):\n",
    "                    upper_id = sequence[id_index + 1]\n",
    "                    upper_hours = recorded_hours[str(upper_id)]\n",
    "                    upper_limits.append(upper_hours)\n",
    "                \n",
    "                # assuming we now have a range of values for the upper and lower limits\n",
    "                # else we'll get division by zero errors\n",
    "                if lower_limits:\n",
    "                    lower_limit = sum(lower_limits)/len(lower_limits)\n",
    "                \n",
    "                if upper_limits:\n",
    "                    upper_limit = sum(upper_limits)/len(upper_limits)\n",
    "                    \n",
    "                \n",
    "                # if we have a situation where we have no upper limits, then we assume that the ticket\n",
    "                # is thought to be bigger than anything else in the system.  \n",
    "                \n",
    "                hours = (lower_limit + upper_limit) / 2\n",
    "    \n",
    "    \n",
    "    return hours\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = prediction_thingy(8, sequences_lists, recorded_hours)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17.633333333333333\n"
     ]
    }
   ],
   "source": [
    "print(prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}