result-categories.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:3594d1213bf6cfec209668f7bf827f71fa59fd942cab93eeedd446ce810aa9a8"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import json\n",
      "import matplotlib.pyplot as plt\n",
      "%matplotlib inline\n",
      "import seaborn as sb\n",
      "import numpy as np\n",
      "import pandas as pd\n",
      "from pprint import pprint\n",
      "from datetime import datetime"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "with open('results-latest.json', 'r') as r:\n",
      "    results = json.load(r)['results']\n",
      "with open('profiles-latest.json', 'r') as p:\n",
      "    profiles = json.load(p)['profiles']\n",
      "print \"{} results loaded\".format(len(results))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "23460 results loaded\n"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "minAppVersionCode = 50\n",
      "parametersVersion = '3.2-production-4'\n",
      "min_n_results = 10\n",
      "good_profiles = [p for p in profiles\n",
      "                 if p['profile_data'].get('appVersionCode') >= minAppVersionCode\n",
      "                 and '+' not in p['profile_data'].get('appVersionName')\n",
      "                 and p['n_results'] >= min_n_results\n",
      "                 and p['profile_data'].get('parametersVersion') == parametersVersion]\n",
      "good_profile_ids = [p['id'] for p in good_profiles]\n",
      "print (\"{n_profiles} profiles worth keeping (>= {min_n_results} results, \"\n",
      "       \"appVersionCode >= {minAppVersionCode}, parametersVersion == {parametersVersion})\").format(\n",
      "            n_profiles=len(good_profiles), min_n_results=min_n_results, \n",
      "            minAppVersionCode=minAppVersionCode, parametersVersion=parametersVersion)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "6 profiles worth keeping (>= 10 results, appVersionCode >= 50, parametersVersion == 3.2-production-4)\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "good_results = [r for r in results if r['profile_id'] in good_profile_ids]\n",
      "print \"{} results uploaded by those good profiles\".format(len(good_results))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "2211 results uploaded by those good profiles\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "ordered_profiles = sorted(good_profiles, key=lambda pr: pr['n_results'], reverse=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "isodate = r'%Y-%m-%dT%H:%M:%S.%fZ'\n",
      "profile_durations = []\n",
      "profile_probe_frequencies = []\n",
      "profile_mornings = []\n",
      "profile_evenings = []\n",
      "\n",
      "for p in ordered_profiles:\n",
      "    n_processed = 0\n",
      "    types = {}\n",
      "    dates = []\n",
      "    for r in good_results:\n",
      "        if r['profile_id'] != p['id']:\n",
      "            continue\n",
      "        \n",
      "        dates.append(r['created_at'])\n",
      "        rdata = r['result_data']\n",
      "        \n",
      "        # Got a location\n",
      "        if 'locationLatitude' in rdata:\n",
      "            if 'location' not in types:\n",
      "                types['location'] = 0\n",
      "            types['location'] += 1\n",
      "        \n",
      "        # Got a probe/questionnaire\n",
      "        elif 'pageGroups' in rdata:\n",
      "            tipe = rdata['type']\n",
      "            status = rdata['status']\n",
      "            if tipe not in types:\n",
      "                types[tipe] = {}\n",
      "            if status not in types[tipe]:\n",
      "                types[tipe][status] = 0\n",
      "            types[tipe][status] += 1\n",
      "        n_processed += 1\n",
      "\n",
      "    \n",
      "    title = \"## Profile {} ({}/{} results processed)\".format(p['id'][:10] + '...', n_processed, p['n_results'])\n",
      "    print '-' * len(title)\n",
      "    print title\n",
      "    print\n",
      "    \n",
      "    min_date = min(dates)\n",
      "    max_date = max(dates)\n",
      "    span_hours = (datetime.strptime(max_date, isodate)\n",
      "                 - datetime.strptime(min_date, isodate)).total_seconds() / 3600.0\n",
      "    span_days = span_hours / 24.0\n",
      "    profile_durations.append(span_days)\n",
      "    profile_probe_frequencies.append(sum(types.get('probe', {}).values()) / span_days)\n",
      "    profile_mornings.append(sum(types.get('morningQuestionnaire', {}).values()))\n",
      "    profile_evenings.append(sum(types.get('eveningQuestionnaire', {}).values()))\n",
      "    \n",
      "    print \" # From {} to {} ({} days)\".format(min(dates)[:10], max(dates)[:10], round(span_days, 1))\n",
      "    print\n",
      "    for tipe, statuses in sorted(types.iteritems(), key=lambda item: item[0]):\n",
      "        if isinstance(statuses, int):\n",
      "            print \"   {}: {} (~{} per day / ~{} per hour)\".format(tipe, statuses, \n",
      "                                                                  round(statuses / span_days, 1),\n",
      "                                                                  round(statuses / span_hours, 1))\n",
      "        else:\n",
      "            total = sum(statuses.values())\n",
      "            print \"   {}: {} (~{} per day)\".format(tipe, total, round(total / span_days, 1))\n",
      "            for status, n in statuses.iteritems():\n",
      "                print \"     {}: {}\".format(status, n)\n",
      "        print\n",
      "    print\n",
      "    print\n",
      "profile_durations = np.array(profile_durations)\n",
      "profile_probe_frequencies = np.array(profile_probe_frequencies)\n",
      "profile_mornings = np.array(profile_mornings)\n",
      "profile_evenings = np.array(profile_evenings)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "------------------------------------------------------\n",
        "## Profile 5705d3b60d... (1902/1902 results processed)\n",
        "\n",
        " # From 2014-11-22 to 2014-12-08 (16.8 days)\n",
        "\n",
        "   beginQuestionnaire: 3 (~0.2 per day)\n",
        "     completed: 3\n",
        "\n",
        "   eveningQuestionnaire: 1 (~0.1 per day)\n",
        "     completed: 1\n",
        "\n",
        "   location: 1892 (~112.6 per day / ~4.7 per hour)\n",
        "\n",
        "   morningQuestionnaire: 1 (~0.1 per day)\n",
        "     completed: 1\n",
        "\n",
        "   probe: 5 (~0.3 per day)\n",
        "     completed: 5\n",
        "\n",
        "\n",
        "\n",
        "----------------------------------------------------\n",
        "## Profile 63da356d30... (165/165 results processed)\n",
        "\n",
        " # From 2014-12-03 to 2014-12-04 (1.4 days)\n",
        "\n",
        "   beginQuestionnaire: 3 (~2.1 per day)\n",
        "     completed: 3\n",
        "\n",
        "   eveningQuestionnaire: 1 (~0.7 per day)\n",
        "     missedOrDismissedOrIncomplete: 1\n",
        "\n",
        "   location: 157 (~110.8 per day / ~4.6 per hour)\n",
        "\n",
        "   morningQuestionnaire: 2 (~1.4 per day)\n",
        "     completed: 2\n",
        "\n",
        "   probe: 2 (~1.4 per day)\n",
        "     completed: 2\n",
        "\n",
        "\n",
        "\n",
        "--------------------------------------------------\n",
        "## Profile daddded666... (67/67 results processed)\n",
        "\n",
        " # From 2014-12-08 to 2014-12-10 (1.9 days)\n",
        "\n",
        "   beginQuestionnaire: 3 (~1.6 per day)\n",
        "     completed: 3\n",
        "\n",
        "   eveningQuestionnaire: 2 (~1.0 per day)\n",
        "     completed: 2\n",
        "\n",
        "   location: 38 (~19.9 per day / ~0.8 per hour)\n",
        "\n",
        "   morningQuestionnaire: 3 (~1.6 per day)\n",
        "     completed: 3\n",
        "\n",
        "   probe: 21 (~11.0 per day)\n",
        "     completed: 21\n",
        "\n",
        "\n",
        "\n",
        "--------------------------------------------------\n",
        "## Profile 966fe7ddf2... (43/43 results processed)\n",
        "\n",
        " # From 2014-12-04 to 2014-12-09 (5.2 days)\n",
        "\n",
        "   beginQuestionnaire: 3 (~0.6 per day)\n",
        "     completed: 3\n",
        "\n",
        "   eveningQuestionnaire: 7 (~1.4 per day)\n",
        "     completed: 5\n",
        "     missedOrDismissedOrIncomplete: 2\n",
        "\n",
        "   morningQuestionnaire: 4 (~0.8 per day)\n",
        "     completed: 4\n",
        "\n",
        "   probe: 29 (~5.6 per day)\n",
        "     completed: 27\n",
        "     missedOrDismissedOrIncomplete: 2\n",
        "\n",
        "\n",
        "\n",
        "--------------------------------------------------\n",
        "## Profile 4a6095da85... (22/22 results processed)\n",
        "\n",
        " # From 2014-11-29 to 2014-11-29 (0.4 days)\n",
        "\n",
        "   beginQuestionnaire: 1 (~2.7 per day)\n",
        "     completed: 1\n",
        "\n",
        "   location: 21 (~57.4 per day / ~2.4 per hour)\n",
        "\n",
        "\n",
        "\n",
        "--------------------------------------------------\n",
        "## Profile 3dd6e7cb67... (12/12 results processed)\n",
        "\n",
        " # From 2014-11-30 to 2014-11-30 (0.2 days)\n",
        "\n",
        "   location: 12 (~71.7 per day / ~3.0 per hour)\n",
        "\n",
        "\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, axs = plt.subplots(2, 2, sharex=True, figsize=(12, 8))\n",
      "palette = sb.color_palette()\n",
      "\n",
      "axs[0, 0].set_axis_off()\n",
      "\n",
      "axs[1, 0].scatter(profile_durations, profile_probe_frequencies, color=palette[0])\n",
      "axs[1, 0].set_title(\"#probes / day vs. #days the profile has participated\")\n",
      "axs[1, 0].set_xlabel(\"#days participated\")\n",
      "axs[1, 0].set_ylabel(\"#probes / day\")\n",
      "\n",
      "axs[0, 1].scatter(profile_durations, profile_mornings, color=palette[1])\n",
      "axs[0, 1].plot([profile_durations.min(), profile_durations.max()], \n",
      "               [profile_durations.min(), profile_durations.max()],\n",
      "               '--', color=palette[1] + (0.3,))\n",
      "axs[0, 1].set_title(\"#morningQuestionnaires vs. #days the profile has participated\")\n",
      "axs[0, 1].set_ylabel(\"#morningQuestionnaires\")\n",
      "\n",
      "axs[1, 1].scatter(profile_durations, profile_evenings, color=palette[2])\n",
      "axs[1, 1].plot([profile_durations.min(), profile_durations.max()], \n",
      "               [profile_durations.min(), profile_durations.max()],\n",
      "               '--', color=palette[2] + (0.3,))\n",
      "axs[1, 1].set_title(\"#eveningQuestionnaires vs. #days the profile has participated\")\n",
      "axs[1, 1].set_xlabel(\"#days participated\")\n",
      "axs[1, 1].set_ylabel(\"#eveningQuestionnaires\")\n",
      "\n",
      "plt.tight_layout()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA1EAAAI5CAYAAABAVbApAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7fa1cb7d5f50>"
       ]
      }
     ],
     "prompt_number": 7
    }
   ],
   "metadata": {}
  }
 ]
}