festival_of_data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 227,
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from shapely.geometry import Point\n",
    "import pandas as pd\n",
    "import geopandas as gpd\n",
    "import networkx as nx\n",
    "from scipy.spatial.distance import cdist\n",
    "import os\n",
    "\n",
    "from citychef import spatial\n",
    "from citychef import graph\n",
    "from citychef import household as hh\n",
    "from citychef import person\n",
    "from citychef import tree\n",
    "from citychef import choice\n",
    "from citychef import osm, gtfs\n",
    "from citychef.household import minmax"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.294285Z",
     "start_time": "2021-09-09T21:27:08.416242Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "source": [
    "city_dir = \"./festival_of_data/city_B\""
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.304717Z",
     "start_time": "2021-09-09T21:27:08.417Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "source": [
    "if not os.path.exists(city_dir):\n",
    "    os.mkdir(city_dir)\n",
    "else:\n",
    "    print(\"WARNING - potentially overwritting previous results\")"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "WARNING - potentially overwritting previous results\n"
     ]
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.307425Z",
     "start_time": "2021-09-09T21:27:08.422Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "source": [
    "# city A\n",
    "# TARGET_HHS = 2000\n",
    "# SEED_AREA = [[275000,780000],[277000,782000]]\n",
    "# TARGET_CENTRES = 5\n",
    "# BUS_ROUTES = 12\n",
    "# TRAIN_ROUTES = 4\n",
    "\n",
    "# city B\n",
    "TARGET_HHS = 2000\n",
    "SEED_AREA = [[278000,783000],[283000,788000]]\n",
    "TARGET_CENTRES = 8\n",
    "BUS_ROUTES = 8\n",
    "TRAIN_ROUTES = 1"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.310727Z",
     "start_time": "2021-09-09T21:27:08.425Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "source": [
    "(x0, y0), (x1, y1) = SEED_AREA\n",
    "area = (x1-x0) * (y1-y0)\n",
    "centres = spatial.Centres(SEED_AREA, density = TARGET_CENTRES/area)\n",
    "\n",
    "spread = ((x1-x0) + (y1-y1))/10\n",
    "facilities = {\n",
    "    'households': spatial.Clusters(centres, size=TARGET_HHS, sigma=spread).crop_to_bbox(SEED_AREA),\n",
    "    'work': spatial.Clusters(centres, size=TARGET_HHS/10, sigma=spread/2).crop_to_bbox(SEED_AREA),\n",
    "    'leisure': spatial.Clusters(centres, size=TARGET_HHS/20, sigma=spread/2).crop_to_bbox(SEED_AREA),\n",
    "    'education': spatial.Clusters(centres, size=TARGET_HHS/20, sigma=spread/3).crop_to_bbox(SEED_AREA),\n",
    "    'health': spatial.Clusters(centres, size=TARGET_HHS/50, sigma=spread/3).crop_to_bbox(SEED_AREA),\n",
    "    'shopping': spatial.Clusters(centres, size=TARGET_HHS/20, sigma=spread/4).crop_to_bbox(SEED_AREA),\n",
    "}\n",
    "bbox = spatial.collect_bbox(facilities) # adjust the bbox for max extends\n",
    "\n",
    "spatial.write_buildings_geojson(\n",
    "    facilities,\n",
    "    path=os.path.join(city_dir, \"building_locations.geojson\"),\n",
    "    epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    "    )"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.315918Z",
     "start_time": "2021-09-09T21:27:08.428Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "source": [
    "car_network = graph.TreeNetwork(bbox, facilities['households'], grid='regular', max_points=50, label=\"highway\")\n",
    "\n",
    "# osm.nx_to_osm(\n",
    "#     g=car_network.g,\n",
    "#     path=os.path.join(city_dir, \"car_osm.xml\")\n",
    "# )\n",
    "graph.nx_to_geojson(\n",
    "    g=car_network.g,\n",
    "    path=os.path.join(city_dir, \"car_network.geojson\"),\n",
    "    epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    ")\n",
    "# fig.savefig(os.path.join(city_dir, 'road_network.png'))"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.318755Z",
     "start_time": "2021-09-09T21:27:08.432Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "source": [
    "buses = graph.Transit(car_network, facilities['households'], density_radius=1000)\n",
    "bus_routes = buses.build_routes(num_routes=BUS_ROUTES, max_length=30, min_length=10, straightness=2)\n",
    "total_bus_graph = buses.graph\n",
    "\n",
    "gtfs.build_gtfs(\n",
    "    transit=buses,\n",
    "    name=\"bus\",\n",
    "    out_dir=city_dir,\n",
    "    agency_id=0,\n",
    "    agency_name='bus_connect',\n",
    "    frequency=10,\n",
    "    from_epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    ")\n",
    "\n",
    "graph.nx_to_geojson(\n",
    "    g=buses.graph,\n",
    "    path=os.path.join(city_dir, \"bus_network.geojson\"),\n",
    "    epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    ")"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "8bfe30ef6e73466eaeba0c5c9e836bc4"
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.321947Z",
     "start_time": "2021-09-09T21:27:08.436Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "source": [
    "potential_stations = np.array([v['pos'] for k, v in car_network.g.nodes.items() if k[:2] == '00'])\n",
    "rail_network = graph.DelaunayNetwork(potential_stations)\n",
    "\n",
    "trains = graph.Transit(rail_network, facilities['households'])\n",
    "train_routes = trains.build_routes(num_routes=TRAIN_ROUTES, max_length=12, min_length=10, straightness=20)\n",
    "\n",
    "total_train_graph = trains.graph\n",
    "\n",
    "gtfs.build_gtfs(\n",
    "    transit=trains,\n",
    "    name=\"rail\",\n",
    "    out_dir=city_dir,\n",
    "    agency_id=1,\n",
    "    agency_name='rail_get_you_there',\n",
    "    frequency=30,\n",
    "    from_epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    ")\n",
    "\n",
    "graph.nx_to_geojson(\n",
    "    g=trains.graph,\n",
    "    path=os.path.join(city_dir, \"rail_network.geojson\"),\n",
    "    epsg=\"epsg:27700\",\n",
    "    to_epsg=\"epsg:4326\"\n",
    ")"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "60ffa50a877f48e2b8fc8589315a6442"
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.324698Z",
     "start_time": "2021-09-09T21:27:08.439Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "source": [
    "fig, ax = plt.subplots(1, 1, figsize=(12, 12))\n",
    "spatial.plot_facilities(facilities, centres=centres, ax=ax)\n",
    "car_network.plot(ax=ax)\n",
    "buses.plot(ax=ax)\n",
    "trains.plot(ax=ax, line_colour='blue')\n",
    "fig.patch.set_visible(False)\n",
    "# ax.axis('off')\n",
    "# plt.axis('equal')\n",
    "\n",
    "# fig.savefig(os.path.join(city_dir, 'city.png'))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/networkx/drawing/nx_pylab.py:579: MatplotlibDeprecationWarning: \n",
      "The iterable function was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use np.iterable instead.\n",
      "  if not cb.iterable(width):\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.329999Z",
     "start_time": "2021-09-09T21:27:08.441Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "source": [
    "hh_df = pd.DataFrame(range(facilities['households'].size), columns=['hh_index'])\n",
    "hh_hidden_df = pd.DataFrame(range(facilities['households'].size), columns=['hh_index'])"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.332660Z",
     "start_time": "2021-09-09T21:27:08.444Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "source": [
    "zones = tree.Zones(\n",
    "    bbox=bbox, \n",
    "    facilities=facilities['households'],\n",
    "    max_zone_facilities=400,\n",
    "    max_sub_zone_facilities=100,\n",
    ")\n",
    "zones.plot()\n",
    "\n",
    "zones.zone_gdf.crs = \"EPSG:27700\"\n",
    "zones.zone_gdf.to_crs(\"epsg:4226\").to_file(os.path.join(city_dir, \"administrative_areas.geojson\"), driver='GeoJSON')\n",
    "\n",
    "zones.sub_zone_gdf.crs = \"EPSG:27700\"\n",
    "zones.sub_zone_gdf.to_crs(\"epsg:4226\").to_file(os.path.join(city_dir, \"administrative_zones.geojson\"), driver='GeoJSON')\n"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x648 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.335305Z",
     "start_time": "2021-09-09T21:27:08.448Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "source": [
    "# fig.savefig(os.path.join(city_dir, 'density.png'))"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.339153Z",
     "start_time": "2021-09-09T21:27:08.451Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "source": [
    "hhs = facilities['households']\n",
    "work = facilities['work']\n",
    "leisure = facilities['leisure']\n",
    "education = facilities['education']\n",
    "health = facilities['health']\n",
    "shopping = facilities['shopping']\n",
    "\n",
    "hh_dist_to_centre = hhs.dist_to_centres()\n",
    "hh_dist_to_centre_mm = minmax(hh_dist_to_centre)\n",
    "\n",
    "hh_df['area_id'] = zones.facility_zone_ids\n",
    "hh_df['zone_id'] = zones.facility_sub_zone_ids\n",
    "\n",
    "n = centres.size\n",
    "\n",
    "hh_hidden_df['density_mm'] = spatial.density(hhs, hhs)\n",
    "hh_hidden_df['dist_closest_centre_mm'] = spatial.distances_to_closest(hhs, centres, 1)\n",
    "hh_hidden_df['dist_closest_centres_mm'] = spatial.distances_to_closest(hhs, centres, max([n,3]))\n",
    "hh_hidden_df['density_work_places_mm'] = spatial.density(hhs, work, 500)\n",
    "hh_hidden_df['density_leisure_mm'] = spatial.density(hhs, leisure, 500)\n",
    "hh_hidden_df['dist_closest_schools_mm'] = spatial.distances_to_closest(hhs, education, max([n,2]))\n",
    "hh_hidden_df['dist_closest_health_mm'] = spatial.density(hhs, health, 500)\n",
    "\n",
    "hh_locs_mm = minmax(hhs.locs)\n",
    "hh_centre_ids_mm = minmax(hhs.ids)\n",
    "\n",
    "hh_hidden_df['hidden'] = hh.gen_hidden(hh_locs_mm[:,0], hh_locs_mm[:,1], hh_hidden_df['density_mm'])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/fred.shone/Projects/citychef/citychef/spatial.py:309: RuntimeWarning: invalid value encountered in true_divide\n",
      "  return (array - min(array)) / (max(array) - min(array))\n",
      "/Users/fred.shone/Projects/citychef/citychef/household.py:26: RuntimeWarning: invalid value encountered in true_divide\n",
      "  choice = choice/sum(choice)\n",
      "/Users/fred.shone/Projects/citychef/citychef/household.py:27: RuntimeWarning: invalid value encountered in less\n",
      "  return np.random.choice([1, 2, 3, 4, 5], p=choice)\n"
     ]
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.342146Z",
     "start_time": "2021-09-09T21:27:08.453Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "source": [
    "hh_hidden_df.head(10)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>density_mm</th>\n",
       "      <th>dist_closest_centre_mm</th>\n",
       "      <th>dist_closest_centres_mm</th>\n",
       "      <th>density_work_places_mm</th>\n",
       "      <th>density_leisure_mm</th>\n",
       "      <th>dist_closest_schools_mm</th>\n",
       "      <th>dist_closest_health_mm</th>\n",
       "      <th>hidden</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.443261</td>\n",
       "      <td>0.370875</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.324795</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.048240</td>\n",
       "      <td>0.076744</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.043173</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.113461</td>\n",
       "      <td>0.240182</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.636364</td>\n",
       "      <td>0.117111</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.332875</td>\n",
       "      <td>0.227630</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.409091</td>\n",
       "      <td>0.204973</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.413595</td>\n",
       "      <td>0.577337</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.045455</td>\n",
       "      <td>0.371968</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.132141</td>\n",
       "      <td>0.149925</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.050563</td>\n",
       "      <td>0.733333</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.270605</td>\n",
       "      <td>0.026839</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.409091</td>\n",
       "      <td>0.206331</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.450684</td>\n",
       "      <td>0.591024</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.385662</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.328482</td>\n",
       "      <td>0.005147</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.345048</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.126099</td>\n",
       "      <td>0.192528</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.123720</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  density_mm  dist_closest_centre_mm  dist_closest_centres_mm  \\\n",
       "0         0         NaN                0.443261                 0.370875   \n",
       "1         1         NaN                0.048240                 0.076744   \n",
       "2         2         NaN                0.113461                 0.240182   \n",
       "3         3         NaN                0.332875                 0.227630   \n",
       "4         4         NaN                0.413595                 0.577337   \n",
       "5         5         NaN                0.132141                 0.149925   \n",
       "6         6         NaN                0.270605                 0.026839   \n",
       "7         7         NaN                0.450684                 0.591024   \n",
       "8         8         NaN                0.328482                 0.005147   \n",
       "9         9         NaN                0.126099                 0.192528   \n",
       "\n",
       "   density_work_places_mm  density_leisure_mm  dist_closest_schools_mm  \\\n",
       "0                    0.10            0.181818                 0.324795   \n",
       "1                    0.98            0.863636                 0.043173   \n",
       "2                    0.66            0.636364                 0.117111   \n",
       "3                    0.40            0.409091                 0.204973   \n",
       "4                    0.06            0.045455                 0.371968   \n",
       "5                    0.88            0.863636                 0.050563   \n",
       "6                    0.58            0.409091                 0.206331   \n",
       "7                    0.06            0.000000                 0.385662   \n",
       "8                    0.26            0.272727                 0.345048   \n",
       "9                    0.38            0.272727                 0.123720   \n",
       "\n",
       "   dist_closest_health_mm  hidden  \n",
       "0                0.000000       1  \n",
       "1                0.933333       1  \n",
       "2                0.466667       1  \n",
       "3                0.200000       4  \n",
       "4                0.000000       4  \n",
       "5                0.733333       4  \n",
       "6                0.400000       5  \n",
       "7                0.000000       4  \n",
       "8                0.133333       1  \n",
       "9                0.133333       5  "
      ]
     },
     "metadata": {},
     "execution_count": 240
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.346721Z",
     "start_time": "2021-09-09T21:27:08.456Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Household Features\n",
    "\n",
    "Household features are generated based on hidden features and previously generated features. All generative funtions combine some level of probabalitic sampling. Data sampling is designed to generate non linear and complex dependancies. The aim of which is to create a population with complex and noisy interdependencies between features."
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-02T12:25:02.854695Z",
     "start_time": "2019-12-02T12:25:02.830443Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "source": [
    "hh_df['hh_count'] = hh.gen_hh_count(\n",
    "    hh_hidden_df['hidden'], \n",
    "    hh_hidden_df['density_mm']\n",
    ")\n",
    "\n",
    "hh_df['hh_children'] = hh.gen_num_children(\n",
    "    hh_df['hh_count'],\n",
    "    hh_hidden_df['hidden'],\n",
    "    hh_hidden_df['dist_closest_schools_mm'],\n",
    "    hh_hidden_df['density_leisure_mm']\n",
    ")\n",
    "\n",
    "hh_hidden_df['age_group'] = hh.gen_age_group(\n",
    "    hh_df['hh_children'],\n",
    "    hh_hidden_df['hidden'],\n",
    "    hh_hidden_df['density_mm']\n",
    ")\n",
    "\n",
    "hh_df['hh_people_in_work'] = hh.get_people_in_work(\n",
    "    hh_hidden_df['age_group'],\n",
    "    hh_hidden_df['hidden'],\n",
    "    hh_df['hh_count'],\n",
    "    hh_df['hh_children'],\n",
    "    hh_hidden_df['density_mm']\n",
    ")\n",
    "\n",
    "hh_df['hh_in_work'] = hh.get_in_work(\n",
    "    hh_df['hh_people_in_work']\n",
    ")\n",
    "\n",
    "hh_df['hh_income'] = hh.get_income(\n",
    "    hh_df['hh_in_work'],\n",
    "    hh_df['hh_count'],\n",
    "    hh_df['hh_children'],\n",
    "    hh_hidden_df['density_work_places_mm'],\n",
    "    hh_hidden_df['density_mm']\n",
    ")\n",
    "\n",
    "hh_income_mm = minmax(hh_df['hh_income'])\n",
    "\n",
    "hh_df['hh_cars'] = hh.get_cars(\n",
    "    hh_hidden_df['hidden'],\n",
    "    hh_income_mm,\n",
    "    hh_hidden_df['density_mm'],\n",
    "    hh_df['hh_count'],\n",
    "    hh_df['hh_children']\n",
    ")"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/fred.shone/Projects/citychef/citychef/household.py:42: RuntimeWarning: invalid value encountered in less\n",
      "  return np.random.choice([1, 2, 3, 4, 5], p=choice)\n",
      "/Users/fred.shone/Projects/citychef/citychef/household.py:80: RuntimeWarning: invalid value encountered in less\n",
      "  p=(p_old, 1-p_old)\n"
     ]
    },
    {
     "output_type": "error",
     "ename": "ValueError",
     "evalue": "p is nan",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-241-12fe0b2b351c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     22\u001b[0m     \u001b[0mhh_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'hh_count'\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[1;32m     23\u001b[0m     \u001b[0mhh_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'hh_children'\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[0;32m---> 24\u001b[0;31m     \u001b[0mhh_hidden_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'density_mm'\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     25\u001b[0m )\n\u001b[1;32m     26\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.ve/citychef/lib/python3.7/site-packages/numpy/lib/function_base.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1970\u001b[0m             \u001b[0mvargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0m_n\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_n\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnames\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[1;32m   1971\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1972\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_vectorize_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvargs\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   1973\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1974\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_get_ufunc_and_otypes\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\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[0;32m~/.ve/citychef/lib/python3.7/site-packages/numpy/lib/function_base.py\u001b[0m in \u001b[0;36m_vectorize_call\u001b[0;34m(self, func, args)\u001b[0m\n\u001b[1;32m   2040\u001b[0m             \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\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[1;32m   2041\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2042\u001b[0;31m             \u001b[0mufunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0motypes\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_ufunc_and_otypes\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0margs\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   2043\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2044\u001b[0m             \u001b[0;31m# Convert args to object arrays first\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.ve/citychef/lib/python3.7/site-packages/numpy/lib/function_base.py\u001b[0m in \u001b[0;36m_get_ufunc_and_otypes\u001b[0;34m(self, func, args)\u001b[0m\n\u001b[1;32m   2000\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2001\u001b[0m             \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflat\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0marg\u001b[0m \u001b[0;32min\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2002\u001b[0;31m             \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minputs\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   2003\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2004\u001b[0m             \u001b[0;31m# Performance note: profiling indicates that -- for simple\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Projects/citychef/citychef/household.py\u001b[0m in \u001b[0;36mget_people_in_work\u001b[0;34m(hh_age_group, hh_hidden, hh_count, hh_num_children, hh_density_mm)\u001b[0m\n\u001b[1;32m    102\u001b[0m     \u001b[0mp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mp\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mp\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mpn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    103\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 104\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbinomial\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0madults\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp\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    105\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    106\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mmtrand.pyx\u001b[0m in \u001b[0;36mmtrand.RandomState.binomial\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: p is nan"
     ]
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.349250Z",
     "start_time": "2021-09-09T21:27:08.460Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "hh_df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>50000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>75000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>85000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0         0      2.0      2.3         3            1                  1   \n",
       "1         1      3.0      3.1         3            1                  1   \n",
       "2         2      9.0      9.1         5            3                  0   \n",
       "3         3      9.0      9.1         2            0                  1   \n",
       "4         4      9.0      9.2         4            1                  3   \n",
       "\n",
       "   hh_in_work  hh_income  hh_cars  \n",
       "0           1      50000        0  \n",
       "1           1      75000        0  \n",
       "2           0      23000        1  \n",
       "3           1      35000        0  \n",
       "4           1      85000        3  "
      ]
     },
     "metadata": {},
     "execution_count": 201
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.353135Z",
     "start_time": "2021-09-09T21:27:08.464Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Individual Features\n",
    "\n",
    "Individuals features are generated based on their household hidden features and previously generated features. As per for households, all generative funtions combine some level of probabalitic sampling. Data sampling is designed to generate non linear and complex dependancies. The aim of which is to create a population with complex and noisy interdependencies between features."
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-02T12:25:06.564761Z",
     "start_time": "2019-12-02T12:25:02.981482Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# hh features\n",
    "hh_counts = hh_df['hh_count'].to_numpy()\n",
    "agent_hh_array = np.repeat(hh_df.to_numpy(), hh_counts, axis=0)\n",
    "person_df = pd.DataFrame(agent_hh_array)\n",
    "person_df.columns = hh_df.columns\n",
    "person_df['p_hh_index'] = np.array([i for c in hh_df['hh_count'] for i in range(c)])\n",
    "\n",
    "#hidden vectors\n",
    "agent_hidden_array = np.repeat(hh_hidden_df.to_numpy(), hh_counts, axis=0)\n",
    "person_hidden_df = pd.DataFrame(agent_hidden_array)\n",
    "person_hidden_df.columns = hh_hidden_df.columns\n",
    "person_hidden_df['p_hh_index'] = np.array([i for c in hh_df['hh_count'] for i in range(c)])"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.357563Z",
     "start_time": "2021-09-09T21:27:08.466Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df['adult'] = person.get_is_adult(\n",
    "    person_df['p_hh_index'],\n",
    "    person_df['hh_count'],\n",
    "    person_df['hh_children']\n",
    ")\n",
    "\n",
    "person_df['gender'] = person.get_gender(\n",
    "    person_df['p_hh_index'], \n",
    "    person_df['adult'],\n",
    "    person_df['hh_children'],\n",
    "    person_hidden_df['hidden']\n",
    ")\n",
    "\n",
    "person_df['age'] = person.get_age(\n",
    "    person_hidden_df['hidden'],\n",
    "    person_hidden_df['age_group'],\n",
    "    person_df['adult']\n",
    ")\n",
    "\n",
    "person_df['employment'] = person.employment(\n",
    "    person_df['adult'],\n",
    "    person_df['hh_people_in_work'],\n",
    "    person_df['p_hh_index'],\n",
    "    person_df['age'],\n",
    "    person_hidden_df['hidden'],\n",
    "    person_hidden_df['dist_closest_schools_mm'],\n",
    "    person_hidden_df['density_work_places_mm'],\n",
    "    person_df['hh_income'],\n",
    "    person_hidden_df['density_mm'],\n",
    ")\n",
    "\n",
    "person_df['occupation'] = person.occupation(\n",
    "    person_df['employment'],\n",
    "    person_df['age'],\n",
    "    minmax(person_df['hh_income']),\n",
    ")\n",
    "\n",
    "person_df.head(10)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "      <th>p_hh_index</th>\n",
       "      <th>adult</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>employment</th>\n",
       "      <th>occupation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23000.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23000.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23000.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23000.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "1       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "2       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "3       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "4       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "5       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "6       2.0      9.0      9.1       5.0          3.0                0.0   \n",
       "7       2.0      9.0      9.1       5.0          3.0                0.0   \n",
       "8       2.0      9.0      9.1       5.0          3.0                0.0   \n",
       "9       2.0      9.0      9.1       5.0          3.0                0.0   \n",
       "\n",
       "   hh_in_work  hh_income  hh_cars  p_hh_index  adult  gender   age  \\\n",
       "0         1.0    50000.0      0.0           0      1       1  22.0   \n",
       "1         1.0    50000.0      0.0           1      1       0  19.0   \n",
       "2         1.0    50000.0      0.0           2      0       1   6.0   \n",
       "3         1.0    75000.0      0.0           0      1       1  22.0   \n",
       "4         1.0    75000.0      0.0           1      1       1  22.0   \n",
       "5         1.0    75000.0      0.0           2      0       0  12.0   \n",
       "6         0.0    23000.0      1.0           0      1       1  22.0   \n",
       "7         0.0    23000.0      1.0           1      1       0  22.0   \n",
       "8         0.0    23000.0      1.0           2      0       0   8.0   \n",
       "9         0.0    23000.0      1.0           3      0       1   9.0   \n",
       "\n",
       "   employment  occupation  \n",
       "0           3           1  \n",
       "1           3           1  \n",
       "2           2           0  \n",
       "3           3           1  \n",
       "4           3           1  \n",
       "5           2           0  \n",
       "6           3           1  \n",
       "7           3           1  \n",
       "8           2           0  \n",
       "9           2           0  "
      ]
     },
     "metadata": {},
     "execution_count": 203
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.360771Z",
     "start_time": "2021-09-09T21:27:08.468Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_hidden_df.head(10)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>density_mm</th>\n",
       "      <th>dist_closest_centre_mm</th>\n",
       "      <th>dist_closest_centres_mm</th>\n",
       "      <th>density_work_places_mm</th>\n",
       "      <th>density_leisure_mm</th>\n",
       "      <th>dist_closest_schools_mm</th>\n",
       "      <th>dist_closest_health_mm</th>\n",
       "      <th>hidden</th>\n",
       "      <th>age_group</th>\n",
       "      <th>p_hh_index</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.324359</td>\n",
       "      <td>0.054413</td>\n",
       "      <td>0.369048</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.292767</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.324359</td>\n",
       "      <td>0.054413</td>\n",
       "      <td>0.369048</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.292767</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.324359</td>\n",
       "      <td>0.054413</td>\n",
       "      <td>0.369048</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.292767</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.224389</td>\n",
       "      <td>0.124667</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.201346</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.224389</td>\n",
       "      <td>0.124667</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.201346</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.224389</td>\n",
       "      <td>0.124667</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.201346</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.244266</td>\n",
       "      <td>0.106735</td>\n",
       "      <td>0.488095</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.117110</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.244266</td>\n",
       "      <td>0.106735</td>\n",
       "      <td>0.488095</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.117110</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.244266</td>\n",
       "      <td>0.106735</td>\n",
       "      <td>0.488095</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.117110</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.244266</td>\n",
       "      <td>0.106735</td>\n",
       "      <td>0.488095</td>\n",
       "      <td>0.638889</td>\n",
       "      <td>0.117110</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  density_mm  dist_closest_centre_mm  dist_closest_centres_mm  \\\n",
       "0       0.0         0.0                0.324359                 0.054413   \n",
       "1       0.0         0.0                0.324359                 0.054413   \n",
       "2       0.0         0.0                0.324359                 0.054413   \n",
       "3       1.0         0.0                0.224389                 0.124667   \n",
       "4       1.0         0.0                0.224389                 0.124667   \n",
       "5       1.0         0.0                0.224389                 0.124667   \n",
       "6       2.0         0.0                0.244266                 0.106735   \n",
       "7       2.0         0.0                0.244266                 0.106735   \n",
       "8       2.0         0.0                0.244266                 0.106735   \n",
       "9       2.0         0.0                0.244266                 0.106735   \n",
       "\n",
       "   density_work_places_mm  density_leisure_mm  dist_closest_schools_mm  \\\n",
       "0                0.369048            0.638889                 0.292767   \n",
       "1                0.369048            0.638889                 0.292767   \n",
       "2                0.369048            0.638889                 0.292767   \n",
       "3                0.428571            0.638889                 0.201346   \n",
       "4                0.428571            0.638889                 0.201346   \n",
       "5                0.428571            0.638889                 0.201346   \n",
       "6                0.488095            0.638889                 0.117110   \n",
       "7                0.488095            0.638889                 0.117110   \n",
       "8                0.488095            0.638889                 0.117110   \n",
       "9                0.488095            0.638889                 0.117110   \n",
       "\n",
       "   dist_closest_health_mm  hidden  age_group  p_hh_index  \n",
       "0                0.666667     1.0       18.0           0  \n",
       "1                0.666667     1.0       18.0           1  \n",
       "2                0.666667     1.0       18.0           2  \n",
       "3                0.600000     1.0       19.0           0  \n",
       "4                0.600000     1.0       19.0           1  \n",
       "5                0.600000     1.0       19.0           2  \n",
       "6                0.533333     1.0       20.0           0  \n",
       "7                0.533333     1.0       20.0           1  \n",
       "8                0.533333     1.0       20.0           2  \n",
       "9                0.533333     1.0       20.0           3  "
      ]
     },
     "metadata": {},
     "execution_count": 204
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.364888Z",
     "start_time": "2021-09-09T21:27:08.470Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "print(len(person_df))\n",
    "print(len(hh_df))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "6728\n",
      "1826\n"
     ]
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.367639Z",
     "start_time": "2021-09-09T21:27:08.472Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Activity Choice\n",
    "\n",
    "A simple activity plan is generated based on individual features. This plan consists of a single activity at a given facility location.\n",
    "\n",
    "#### Method\n",
    "1. Activity type choice is made based on household and individual features. This can include staying at home.\n",
    "2. Facilities for each activity type are weighted by their desirability (based on density and distance).\n",
    "3. Agents randomly choose a facility based on this weighting and on their individual features."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# Main activity choice\n",
    "person_df['main_activity'] = choice.main_activity_choice(\n",
    "    person_df['employment'],\n",
    "    person_hidden_df['density_work_places_mm'],\n",
    "    person_df['occupation'],\n",
    "    person_hidden_df['hidden'],\n",
    "    minmax(person_df['hh_income']),\n",
    "    person_df['hh_count'],\n",
    "    person_hidden_df['dist_closest_centres_mm'],\n",
    "    person_hidden_df['density_leisure_mm'],\n",
    "    person_df['age'],\n",
    ")"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.375770Z",
     "start_time": "2021-09-09T21:27:08.474Z"
    },
    "code_folding": []
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "      <th>p_hh_index</th>\n",
       "      <th>adult</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>employment</th>\n",
       "      <th>occupation</th>\n",
       "      <th>main_activity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "1       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "2       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "3       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "4       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "\n",
       "   hh_in_work  hh_income  hh_cars  p_hh_index  adult  gender   age  \\\n",
       "0         1.0    50000.0      0.0           0      1       1  22.0   \n",
       "1         1.0    50000.0      0.0           1      1       0  19.0   \n",
       "2         1.0    50000.0      0.0           2      0       1   6.0   \n",
       "3         1.0    75000.0      0.0           0      1       1  22.0   \n",
       "4         1.0    75000.0      0.0           1      1       1  22.0   \n",
       "\n",
       "   employment  occupation  main_activity  \n",
       "0           3           1              1  \n",
       "1           3           1              5  \n",
       "2           2           0              2  \n",
       "3           3           1              1  \n",
       "4           3           1              1  "
      ]
     },
     "metadata": {},
     "execution_count": 207
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.383013Z",
     "start_time": "2021-09-09T21:27:08.477Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# FACILITY FEATURES\n",
    "zone_gdf = zones.sub_zone_gdf\n",
    "\n",
    "work_df = pd.DataFrame(range(work.size), columns=['index'])\n",
    "work_df['density_mm'] = spatial.density(work, work, 500)\n",
    "work_df['geometry'] = [Point(x,y) for x,y in work.locs]\n",
    "work_gdf = gpd.GeoDataFrame(work_df, geometry='geometry')\n",
    "work_gdf = gpd.sjoin(work_gdf, zone_gdf, how=\"left\", op='within').drop('index_right', axis=1)\n",
    "\n",
    "leisure_df = pd.DataFrame(range(leisure.size), columns=['index'])\n",
    "leisure_df['density_mm'] = spatial.density(leisure, shopping, 500)\n",
    "leisure_df['geometry'] = [Point(x,y) for x,y in leisure.locs]\n",
    "leisure_gdf = gpd.GeoDataFrame(leisure_df, geometry='geometry')\n",
    "leisure_gdf = gpd.sjoin(leisure_gdf, zone_gdf, how=\"left\", op='within').drop('index_right', axis=1)\n",
    "\n",
    "education_df = pd.DataFrame(range(education.size), columns=['index'])\n",
    "education_df['density_mm'] = spatial.density(education, shopping, 500)\n",
    "education_df['geometry'] = [Point(x,y) for x,y in education.locs]\n",
    "education_gdf = gpd.GeoDataFrame(education_df, geometry='geometry')\n",
    "education_gdf = gpd.sjoin(education_gdf, zone_gdf, how=\"left\", op='within').drop('index_right', axis=1)\n",
    "\n",
    "health_df = pd.DataFrame(range(health.size), columns=['index'])\n",
    "health_df['density_mm'] = spatial.density(health, shopping, 500)\n",
    "health_df['geometry'] = [Point(x,y) for x,y in health.locs]\n",
    "health_gdf = gpd.GeoDataFrame(health_df, geometry='geometry')\n",
    "health_gdf = gpd.sjoin(health_gdf, zone_gdf, how=\"left\", op='within').drop('index_right', axis=1)\n",
    "\n",
    "shopping_df = pd.DataFrame(range(shopping.size), columns=['index'])\n",
    "shopping_df['density_mm'] = spatial.density(shopping, shopping, 300)\n",
    "shopping_df['geometry'] = [Point(x,y) for x,y in shopping.locs]\n",
    "shopping_gdf = gpd.GeoDataFrame(shopping_df, geometry='geometry')\n",
    "shopping_gdf = gpd.sjoin(shopping_gdf, zone_gdf, how=\"left\", op='within').drop('index_right', axis=1)\n"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/geopandas/tools/sjoin.py:61: UserWarning: CRS of frames being joined does not match!(None != EPSG:27700)\n",
      "  \"(%s != %s)\" % (left_df.crs, right_df.crs)\n",
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/geopandas/tools/sjoin.py:61: UserWarning: CRS of frames being joined does not match!(None != EPSG:27700)\n",
      "  \"(%s != %s)\" % (left_df.crs, right_df.crs)\n",
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/geopandas/tools/sjoin.py:61: UserWarning: CRS of frames being joined does not match!(None != EPSG:27700)\n",
      "  \"(%s != %s)\" % (left_df.crs, right_df.crs)\n",
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/geopandas/tools/sjoin.py:61: UserWarning: CRS of frames being joined does not match!(None != EPSG:27700)\n",
      "  \"(%s != %s)\" % (left_df.crs, right_df.crs)\n",
      "/Users/fred.shone/.ve/citychef/lib/python3.7/site-packages/geopandas/tools/sjoin.py:61: UserWarning: CRS of frames being joined does not match!(None != EPSG:27700)\n",
      "  \"(%s != %s)\" % (left_df.crs, right_df.crs)\n"
     ]
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.388967Z",
     "start_time": "2021-09-09T21:27:08.479Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "hhs_work_distances = minmax(cdist(hhs.locs, work.locs))\n",
    "hhs_education_distances = minmax(cdist(hhs.locs, education.locs))\n",
    "hhs_shopping_distances = minmax(cdist(hhs.locs, shopping.locs))\n",
    "hhs_leisure_distances = minmax(cdist(hhs.locs, leisure.locs))\n",
    "hhs_health_distances = minmax(cdist(hhs.locs, health.locs))"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.393475Z",
     "start_time": "2021-09-09T21:27:08.481Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "main_facility_id = np.zeros((len(person_df)))\n",
    "main_facility_area = np.zeros((len(person_df)))\n",
    "main_facility_zone = np.zeros((len(person_df)))\n",
    "\n",
    "for i, (hh_index, main_act, income_mm) in enumerate(zip(person_df.hh_index, person_df.main_activity, minmax(person_df['hh_income']))):\n",
    "    \n",
    "    if main_act == 0:  # home\n",
    "        main_facility_id[i] = -1\n",
    "        main_facility_area[i] = -1\n",
    "        main_facility_zone[i] = -1\n",
    "        \n",
    "    hh_index = int(hh_index)\n",
    "        \n",
    "    if main_act == 1:  # work\n",
    "        facility_distances = hhs_work_distances[hh_index]\n",
    "        facility_scores = (1 - income_mm) * (1 - facility_distances) + income_mm * work_df.density_mm\n",
    "        facility_scores = facility_scores / facility_scores.sum()\n",
    "        facitlity_id = np.random.choice(range(len(facility_scores)), p=facility_scores)\n",
    "        main_facility_id[i] = facitlity_id\n",
    "        main_facility_area[i], main_facility_zone[i] = work_gdf[['area_id', 'zone_id']].iloc[facitlity_id]\n",
    "        continue\n",
    "        \n",
    "    if main_act == 2:  # education\n",
    "        facility_distances = hhs_education_distances[hh_index]\n",
    "        facility_scores = (1 - facility_distances) ** 2\n",
    "        facility_scores = facility_scores / facility_scores.sum()\n",
    "        facitlity_id = np.random.choice(range(len(facility_scores)), p=facility_scores)\n",
    "        main_facility_id[i] = facitlity_id\n",
    "        main_facility_area[i], main_facility_zone[i] = education_gdf[['area_id', 'zone_id']].iloc[facitlity_id]\n",
    "        continue\n",
    "        \n",
    "    if main_act == 3:  # shopping\n",
    "        facility_distances = hhs_shopping_distances[hh_index]\n",
    "        facility_scores = (1 - income_mm) * (1 - facility_distances) + income_mm * shopping_df.density_mm\n",
    "        facility_scores = facility_scores / facility_scores.sum()\n",
    "        facitlity_id = np.random.choice(range(len(facility_scores)), p=facility_scores)\n",
    "        main_facility_id[i] = facitlity_id\n",
    "        main_facility_area[i], main_facility_zone[i] = shopping_gdf[['area_id', 'zone_id']].iloc[facitlity_id]\n",
    "        continue\n",
    "        \n",
    "    if main_act == 4:  # leisure\n",
    "        facility_distances = hhs_leisure_distances[hh_index]\n",
    "        facility_scores = (1 - facility_distances)\n",
    "        facility_scores = facility_scores / facility_scores.sum()\n",
    "        facitlity_id = np.random.choice(range(len(facility_scores)), p=facility_scores)\n",
    "        main_facility_id[i] = facitlity_id\n",
    "        main_facility_area[i], main_facility_zone[i] = leisure_gdf[['area_id', 'zone_id']].iloc[facitlity_id]\n",
    "        continue\n",
    "        \n",
    "    if main_act == 5:  # health\n",
    "        facility_distances = hhs_health_distances[hh_index]\n",
    "        facility_scores = (1 - facility_distances) ** 4\n",
    "        facility_scores = facility_scores / facility_scores.sum()\n",
    "        facitlity_id = np.random.choice(range(len(facility_scores)), p=facility_scores)\n",
    "        main_facility_id[i] = facitlity_id\n",
    "        main_facility_area[i], main_facility_zone[i] = health_gdf[['area_id', 'zone_id']].iloc[facitlity_id]\n",
    "        continue\n",
    "\n",
    "person_df['main_activity_id'] = main_facility_id\n",
    "person_df['main_facility_area'] = main_facility_area\n",
    "person_df['main_facility_zone'] = main_facility_zone\n"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.401993Z",
     "start_time": "2021-09-09T21:27:08.482Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# HH CAR CONNECTIVITY\n",
    "car_ods = graph.NodesOD(car_network.g)\n",
    "\n",
    "hh_df['car_dist'], hh_df['car_node'] = spatial.distance_index_nearest_node(hhs, car_network.locs)\n",
    "work_df['car_dist'], work_df['car_node'] = spatial.distance_index_nearest_node(work, car_network.locs)\n",
    "leisure_df['car_dist'], leisure_df['car_node'] = spatial.distance_index_nearest_node(leisure, car_network.locs)\n",
    "education_df['car_dist'], education_df['car_node'] = spatial.distance_index_nearest_node(education, car_network.locs)\n",
    "health_df['car_dist'], health_df['car_node'] = spatial.distance_index_nearest_node(health, car_network.locs)\n",
    "shopping_df['car_dist'], shopping_df['car_node'] = spatial.distance_index_nearest_node(shopping, car_network.locs)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.406098Z",
     "start_time": "2021-09-09T21:27:08.484Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "hh_df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "      <th>car_dist</th>\n",
       "      <th>car_node</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>50000</td>\n",
       "      <td>0</td>\n",
       "      <td>45.512609</td>\n",
       "      <td>171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>75000</td>\n",
       "      <td>0</td>\n",
       "      <td>43.025036</td>\n",
       "      <td>157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23000</td>\n",
       "      <td>1</td>\n",
       "      <td>48.870435</td>\n",
       "      <td>535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35000</td>\n",
       "      <td>0</td>\n",
       "      <td>79.092904</td>\n",
       "      <td>539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>85000</td>\n",
       "      <td>3</td>\n",
       "      <td>58.305410</td>\n",
       "      <td>529</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0         0      2.0      2.3         3            1                  1   \n",
       "1         1      3.0      3.1         3            1                  1   \n",
       "2         2      9.0      9.1         5            3                  0   \n",
       "3         3      9.0      9.1         2            0                  1   \n",
       "4         4      9.0      9.2         4            1                  3   \n",
       "\n",
       "   hh_in_work  hh_income  hh_cars   car_dist  car_node  \n",
       "0           1      50000        0  45.512609       171  \n",
       "1           1      75000        0  43.025036       157  \n",
       "2           0      23000        1  48.870435       535  \n",
       "3           1      35000        0  79.092904       539  \n",
       "4           1      85000        3  58.305410       529  "
      ]
     },
     "metadata": {},
     "execution_count": 212
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.409035Z",
     "start_time": "2021-09-09T21:27:08.486Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# HH PT CONNECTIVITY\n",
    "pt_network = nx.compose(total_bus_graph, total_train_graph)\n",
    "pt_pos = {n:d['pos'] for n,d in pt_network.nodes(data=True)}\n",
    "pt_node_locs = np.array([d['pos'] for n,d in pt_network.nodes(data=True)])\n",
    "pt_ods = graph.NodesOD(pt_network)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.417133Z",
     "start_time": "2021-09-09T21:27:08.488Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "hh_df['pt_dist'], hh_df['pt_node'] = spatial.distance_index_nearest_node(hhs, pt_node_locs)\n",
    "work_df['pt_dist'], work_df['pt_node'] = spatial.distance_index_nearest_node(work, pt_node_locs)\n",
    "leisure_df['pt_dist'], leisure_df['pt_node'] = spatial.distance_index_nearest_node(leisure, pt_node_locs)\n",
    "education_df['pt_dist'], education_df['pt_node'] = spatial.distance_index_nearest_node(education, pt_node_locs)\n",
    "health_df['pt_dist'], health_df['pt_node'] = spatial.distance_index_nearest_node(health, pt_node_locs)\n",
    "shopping_df['pt_dist'], shopping_df['pt_node'] = spatial.distance_index_nearest_node(shopping, pt_node_locs)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.421377Z",
     "start_time": "2021-09-09T21:27:08.490Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "hh_df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "      <th>car_dist</th>\n",
       "      <th>car_node</th>\n",
       "      <th>pt_dist</th>\n",
       "      <th>pt_node</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>50000</td>\n",
       "      <td>0</td>\n",
       "      <td>45.512609</td>\n",
       "      <td>171</td>\n",
       "      <td>65.990680</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>75000</td>\n",
       "      <td>0</td>\n",
       "      <td>43.025036</td>\n",
       "      <td>157</td>\n",
       "      <td>81.182667</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23000</td>\n",
       "      <td>1</td>\n",
       "      <td>48.870435</td>\n",
       "      <td>535</td>\n",
       "      <td>167.901605</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35000</td>\n",
       "      <td>0</td>\n",
       "      <td>79.092904</td>\n",
       "      <td>539</td>\n",
       "      <td>133.986423</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>85000</td>\n",
       "      <td>3</td>\n",
       "      <td>58.305410</td>\n",
       "      <td>529</td>\n",
       "      <td>58.305410</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0         0      2.0      2.3         3            1                  1   \n",
       "1         1      3.0      3.1         3            1                  1   \n",
       "2         2      9.0      9.1         5            3                  0   \n",
       "3         3      9.0      9.1         2            0                  1   \n",
       "4         4      9.0      9.2         4            1                  3   \n",
       "\n",
       "   hh_in_work  hh_income  hh_cars   car_dist  car_node     pt_dist  pt_node  \n",
       "0           1      50000        0  45.512609       171   65.990680        8  \n",
       "1           1      75000        0  43.025036       157   81.182667       11  \n",
       "2           0      23000        1  48.870435       535  167.901605       24  \n",
       "3           1      35000        0  79.092904       539  133.986423       27  \n",
       "4           1      85000        3  58.305410       529   58.305410       23  "
      ]
     },
     "metadata": {},
     "execution_count": 215
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.431032Z",
     "start_time": "2021-09-09T21:27:08.492Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "car_time = np.zeros((len(person_df)))\n",
    "pt_time = np.zeros((len(person_df)))\n",
    "cycle_time = np.zeros((len(person_df)))\n",
    "walk_time = np.zeros((len(person_df)))\n",
    "mode_choice = np.zeros((len(person_df)))\n",
    "journey_time = np.zeros((len(person_df)))\n",
    "\n",
    "for i, (hh_index, main_act, main_id, income_mm, age_mm, adult, children, cars) in enumerate(\n",
    "    zip(\n",
    "        person_df.hh_index,\n",
    "        person_df.main_activity,\n",
    "        person_df.main_activity_id,\n",
    "        minmax(person_df.hh_income),\n",
    "        minmax(person_df.age),\n",
    "        person_df.adult,\n",
    "        person_df.hh_children,\n",
    "        person_df.hh_cars,\n",
    "              )):\n",
    "    \n",
    "    if main_act == 0:  # home\n",
    "        mode_choice[i] = -1\n",
    "        continue\n",
    "        \n",
    "    main_id = int(main_id)\n",
    "    hh_index = int(hh_index)\n",
    "    \n",
    "    hh_loc = hhs.locs[hh_index]\n",
    "    \n",
    "    hh_car_node_dist, hh_car_node = hh_df.car_dist[hh_index], hh_df.car_node[hh_index]\n",
    "    hh_car_node_time = 1.4 * 3600 * hh_car_node_dist / 25\n",
    "    \n",
    "    hh_pt_node_dist, hh_pt_node = hh_df.pt_dist[hh_index], hh_df.pt_node[hh_index]\n",
    "    hh_pt_node_time = 3600 * hh_car_node_dist / 5\n",
    "    \n",
    "    times = {}\n",
    "    options = {}\n",
    "        \n",
    "    if main_act == 1:  # work\n",
    "        \n",
    "        facility_loc = work.locs[main_id]\n",
    "        facility_df = work_df\n",
    "        \n",
    "    if main_act == 2:  # education\n",
    "        \n",
    "        facility_loc = education.locs[main_id]\n",
    "        facility_df = education_df\n",
    "        \n",
    "    if main_act == 3:  # shopping\n",
    "        \n",
    "        facility_loc = shopping.locs[main_id]\n",
    "        facility_df = shopping_df\n",
    "        \n",
    "    if main_act == 4:  # leisure\n",
    "        \n",
    "        facility_loc = leisure.locs[main_id]\n",
    "        facility_df = leisure_df\n",
    "        \n",
    "    if main_act == 5:  # health\n",
    "        \n",
    "        facility_loc = health.locs[main_id]\n",
    "        facility_df = health_df\n",
    "     \n",
    "    # choice\n",
    "    eucl_distance = ((hh_loc - facility_loc)**2).sum()**.5\n",
    "    manh_distance = np.abs(hh_loc - facility_loc).sum()\n",
    "    \n",
    "    # car\n",
    "    facility_car_node_dist, facility_car_node = facility_df.car_dist[main_id], facility_df.car_node[main_id]\n",
    "    car_network_travel = car_ods.get(hh_car_node, facility_car_node)\n",
    "    if car_network_travel != -1:\n",
    "        facility_car_node_time = 1.4 * 3600 * facility_car_node_dist / 25\n",
    "        car_journey_time = hh_car_node_time + facility_car_node_time + car_network_travel + 300\n",
    "        if car_journey_time > 300:\n",
    "            car_time[i] = car_journey_time\n",
    "            times[0] = car_journey_time\n",
    "\n",
    "            if not cars:\n",
    "                options[0] =  3 / car_journey_time\n",
    "            else:\n",
    "                options[0] = 1 / car_journey_time * (1.1 - cars / 10) * (1 - children / 10)\n",
    "    \n",
    "    # pt\n",
    "    facility_pt_node_dist, facility_pt_node = facility_df.pt_dist[main_id], facility_df.pt_node[main_id]\n",
    "    pt_network_travel = pt_ods.get(hh_pt_node, facility_pt_node)\n",
    "    if pt_network_travel != -1:\n",
    "        facility_pt_node_time = 3600 * facility_pt_node_dist / 5\n",
    "        pt_journey_time = hh_pt_node_time + facility_pt_node_time + pt_network_travel + 120\n",
    "        if pt_journey_time > 300:\n",
    "            pt_time[i] = pt_journey_time\n",
    "            times[1] = pt_journey_time\n",
    "\n",
    "            options[1] = 1/ pt_journey_time * (1 + age_mm / 100) * (1 + income_mm / 10)\n",
    "    \n",
    "    # cycle\n",
    "    cycle_journey_time = 360 * manh_distance / 15 + 120\n",
    "    if (cycle_journey_time < 3600) and (cycle_journey_time > 120):\n",
    "        cycle_time[i] = cycle_journey_time\n",
    "        times[2] = cycle_journey_time\n",
    "        \n",
    "        if main_act == 3:\n",
    "            options[2] = 1/cycle_journey_time * 3\n",
    "        else:\n",
    "            options[2] = 1/cycle_journey_time * (1 + age_mm / 10) * (1 + income_mm / 10)\n",
    "\n",
    "    # walk\n",
    "    walk_journey_time = 360 * eucl_distance / 5\n",
    "    if walk_journey_time < 3600:\n",
    "        walk_time[i] = walk_journey_time\n",
    "        times[3] = walk_journey_time\n",
    "        options[3] = 1/walk_journey_time\n",
    "    \n",
    "    weights = np.array(list(options.values()))\n",
    "    weights = weights / weights.sum()\n",
    "    mode = np.random.choice(list(options), p=weights)\n",
    "    time = times[mode]\n",
    "    \n",
    "    mode_choice[i] = mode\n",
    "    journey_time[i] = time\n"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.434612Z",
     "start_time": "2021-09-09T21:27:08.494Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df['mode'] = mode_choice\n",
    "person_hidden_df['journey_time'] = journey_time"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.437293Z",
     "start_time": "2021-09-09T21:27:08.497Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "header = {\n",
    "    'hh_in_work': {\n",
    "        0: 'noone',\n",
    "        1: 'yes',\n",
    "    },\n",
    "    'adult': {\n",
    "        0: 'under16',\n",
    "        1: '16+',\n",
    "    },\n",
    "    'gender': {\n",
    "        0: 'male',\n",
    "        1: 'female',\n",
    "        2: 'other',\n",
    "    },\n",
    "    'employment': {\n",
    "        0: 'retired',\n",
    "        1: 'unemployed',\n",
    "        2: 'education',\n",
    "        3: 'employment',\n",
    "    },\n",
    "    'occupation': {\n",
    "        0: 'none',\n",
    "        1: 'unskilled',\n",
    "        2: 'skilled',\n",
    "        3: 'manager',\n",
    "    },\n",
    "    'main_activity': {\n",
    "        0: 'None',\n",
    "        1: 'work',\n",
    "        2: 'education',\n",
    "        3: 'shopping',\n",
    "        4: 'leisure',\n",
    "        5: 'health',\n",
    "    },\n",
    "    'mode': {\n",
    "        0: 'car',\n",
    "        1: 'PT',\n",
    "        2: 'cycle',\n",
    "        3: 'walk',\n",
    "    },\n",
    "    'hh_income_bin': {\n",
    "        0: 'low',\n",
    "        1: 'mid',\n",
    "        2: 'mid-high',\n",
    "        3: 'high',\n",
    "    },\n",
    "    'age_bin': {\n",
    "        0: '-4',\n",
    "        4: '4-13',\n",
    "        13: '13-16',\n",
    "        16: '16-21',\n",
    "        21: '21-31',\n",
    "        31: '31-41',\n",
    "        41: '41-51',\n",
    "        51: '51-61',\n",
    "        61: '61-71',\n",
    "        71: '71-81',\n",
    "        81: '81+',\n",
    "    }\n",
    "}"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.440451Z",
     "start_time": "2021-09-09T21:27:08.499Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# import json\n",
    "# with open(os.path.join(city_dir, 'header.json'), 'w') as f:\n",
    "#     json.dump(header, f)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.445017Z",
     "start_time": "2021-09-09T21:27:08.504Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df.columns"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "Index(['hh_index', 'area_id', 'zone_id', 'hh_count', 'hh_children',\n",
       "       'hh_people_in_work', 'hh_in_work', 'hh_income', 'hh_cars', 'p_hh_index',\n",
       "       'adult', 'gender', 'age', 'employment', 'occupation', 'main_activity',\n",
       "       'main_activity_id', 'main_facility_area', 'main_facility_zone', 'mode'],\n",
       "      dtype='object')"
      ]
     },
     "metadata": {},
     "execution_count": 220
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.447769Z",
     "start_time": "2021-09-09T21:27:08.509Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# bin the continuous variables (income and age)\n",
    "d = person_df.hh_income.describe()\n",
    "bins = [0, d['25%'], d['50%'], d['75%'], d['max']]\n",
    "labels = [0,1,2,3]\n",
    "person_df['hh_income_bin'] = pd.cut(person_df.hh_income, bins=bins, labels=labels)\n",
    "\n",
    "bins = [-1, 4, 13, 16, 21, 31, 41, 51, 61, 71, 81, 120]\n",
    "labels = ['0-4', '4-13', '13-16','16-21', '21-31', '31-41', '41-51', '51-61', '61-71', '71-81', '81+']\n",
    "# labels = [0, 4, 13, 16, 21, 31, 41, 51, 61, 71, 81]\n",
    "person_df['age_bin'] = pd.cut(person_df.age, bins=bins, labels=labels)\n",
    "\n",
    "person_df.hh_in_work = person_df.hh_in_work.map(header[\"hh_in_work\"])\n",
    "person_df.adult = person_df.adult.map(header[\"adult\"])\n",
    "person_df.gender = person_df.gender.map(header[\"gender\"])\n",
    "person_df.employment = person_df.employment.map(header[\"employment\"])\n",
    "person_df.occupation = person_df.occupation.map(header[\"occupation\"])\n",
    "person_df.main_activity = person_df.main_activity.map(header[\"main_activity\"])\n",
    "person_df[\"mode\"] = person_df[\"mode\"].map(header[\"mode\"])\n",
    "person_df.hh_income_bin = person_df.hh_income_bin.map(header[\"hh_income_bin\"])\n",
    "\n",
    "person_df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hh_index</th>\n",
       "      <th>area_id</th>\n",
       "      <th>zone_id</th>\n",
       "      <th>hh_count</th>\n",
       "      <th>hh_children</th>\n",
       "      <th>hh_people_in_work</th>\n",
       "      <th>hh_in_work</th>\n",
       "      <th>hh_income</th>\n",
       "      <th>hh_cars</th>\n",
       "      <th>p_hh_index</th>\n",
       "      <th>...</th>\n",
       "      <th>age</th>\n",
       "      <th>employment</th>\n",
       "      <th>occupation</th>\n",
       "      <th>main_activity</th>\n",
       "      <th>main_activity_id</th>\n",
       "      <th>main_facility_area</th>\n",
       "      <th>main_facility_zone</th>\n",
       "      <th>mode</th>\n",
       "      <th>hh_income_bin</th>\n",
       "      <th>age_bin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>yes</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>22.0</td>\n",
       "      <td>employment</td>\n",
       "      <td>unskilled</td>\n",
       "      <td>work</td>\n",
       "      <td>204.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>car</td>\n",
       "      <td>mid</td>\n",
       "      <td>21-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>yes</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>19.0</td>\n",
       "      <td>employment</td>\n",
       "      <td>unskilled</td>\n",
       "      <td>health</td>\n",
       "      <td>6.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.3</td>\n",
       "      <td>car</td>\n",
       "      <td>mid</td>\n",
       "      <td>16-21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>yes</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>education</td>\n",
       "      <td>none</td>\n",
       "      <td>education</td>\n",
       "      <td>7.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>car</td>\n",
       "      <td>mid</td>\n",
       "      <td>4-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>yes</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>22.0</td>\n",
       "      <td>employment</td>\n",
       "      <td>unskilled</td>\n",
       "      <td>work</td>\n",
       "      <td>64.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.4</td>\n",
       "      <td>car</td>\n",
       "      <td>mid-high</td>\n",
       "      <td>21-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>yes</td>\n",
       "      <td>75000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>22.0</td>\n",
       "      <td>employment</td>\n",
       "      <td>unskilled</td>\n",
       "      <td>work</td>\n",
       "      <td>8.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.3</td>\n",
       "      <td>car</td>\n",
       "      <td>mid-high</td>\n",
       "      <td>21-31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   hh_index  area_id  zone_id  hh_count  hh_children  hh_people_in_work  \\\n",
       "0       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "1       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "2       0.0      2.0      2.3       3.0          1.0                1.0   \n",
       "3       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "4       1.0      3.0      3.1       3.0          1.0                1.0   \n",
       "\n",
       "  hh_in_work  hh_income  hh_cars  p_hh_index  ...   age  employment  \\\n",
       "0        yes    50000.0      0.0           0  ...  22.0  employment   \n",
       "1        yes    50000.0      0.0           1  ...  19.0  employment   \n",
       "2        yes    50000.0      0.0           2  ...   6.0   education   \n",
       "3        yes    75000.0      0.0           0  ...  22.0  employment   \n",
       "4        yes    75000.0      0.0           1  ...  22.0  employment   \n",
       "\n",
       "   occupation main_activity main_activity_id main_facility_area  \\\n",
       "0   unskilled          work            204.0                0.0   \n",
       "1   unskilled        health              6.0               10.0   \n",
       "2        none     education              7.0               10.0   \n",
       "3   unskilled          work             64.0                5.0   \n",
       "4   unskilled          work              8.0               10.0   \n",
       "\n",
       "   main_facility_zone  mode  hh_income_bin age_bin  \n",
       "0                 0.1   car            mid   21-31  \n",
       "1                10.3   car            mid   16-21  \n",
       "2                10.0   car            mid    4-13  \n",
       "3                 5.4   car       mid-high   21-31  \n",
       "4                10.3   car       mid-high   21-31  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "metadata": {},
     "execution_count": 221
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.450595Z",
     "start_time": "2021-09-09T21:27:08.515Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Commuter Survey Output\n",
    "\n",
    "We extract tours for work and education activities and create an zone based origin - destination freq table.\n",
    "\n",
    "Note that we could further break-down by activity type and or mode choice."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "activities = [\"work\", \"education\"] # work & education\n",
    "commuter_df = person_df.hh_index.loc[person_df.main_activity.isin(activities)]\n",
    "commuter_freq_df = commuter_df.groupby([person_df.zone_id, person_df.main_facility_zone]).count()\n",
    "commuter_freq_df.name = 'freq'\n",
    "commuter_freq_df.to_csv(os.path.join(city_dir, 'zone_commuter_freq_table.csv'), header=True)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.453083Z",
     "start_time": "2021-09-09T21:27:08.517Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "# matrix version\n",
    "commuter_freq_df.unstack().to_csv(os.path.join(city_dir, 'zone_commuter_freq_matrix.csv'), header=True)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.458224Z",
     "start_time": "2021-09-09T21:27:08.518Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df.to_csv(os.path.join(city_dir, 'population_survey.csv'), header=True)"
   ],
   "outputs": [],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.465223Z",
     "start_time": "2021-09-09T21:27:08.520Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "person_df[\"mode\"].value_counts()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "car      5510\n",
       "PT        144\n",
       "cycle      72\n",
       "walk        4\n",
       "Name: mode, dtype: int64"
      ]
     },
     "metadata": {},
     "execution_count": 225
    }
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-09-09T21:27:41.467670Z",
     "start_time": "2021-09-09T21:27:08.523Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "for c in person_df.columns:\n",
    "    print(\"===>\",c)\n",
    "    if len(person_df[c].value_counts())<12:\n",
    "        print(person_df[c].value_counts())"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "===> hh_index\n",
      "===> area_id\n",
      "===> zone_id\n",
      "===> hh_count\n",
      "5.0    3115\n",
      "4.0    1876\n",
      "3.0    1152\n",
      "2.0     470\n",
      "1.0     115\n",
      "Name: hh_count, dtype: int64\n",
      "===> hh_children\n",
      "1.0    2512\n",
      "0.0    2314\n",
      "2.0    1438\n",
      "3.0     414\n",
      "4.0      50\n",
      "Name: hh_children, dtype: int64\n",
      "===> hh_people_in_work\n",
      "2.0    2088\n",
      "3.0    1618\n",
      "1.0    1613\n",
      "0.0     730\n",
      "4.0     654\n",
      "5.0      25\n",
      "Name: hh_people_in_work, dtype: int64\n",
      "===> hh_in_work\n",
      "yes      5998\n",
      "noone     730\n",
      "Name: hh_in_work, dtype: int64\n",
      "===> hh_income\n",
      "===> hh_cars\n",
      "1.0    2518\n",
      "2.0    1822\n",
      "0.0    1507\n",
      "3.0     761\n",
      "4.0      95\n",
      "5.0      25\n",
      "Name: hh_cars, dtype: int64\n",
      "===> p_hh_index\n",
      "0    1826\n",
      "1    1711\n",
      "2    1476\n",
      "3    1092\n",
      "4     623\n",
      "Name: p_hh_index, dtype: int64\n",
      "===> adult\n",
      "16+        5150\n",
      "under16    1578\n",
      "Name: adult, dtype: int64\n",
      "===> gender\n",
      "male      3594\n",
      "female    2988\n",
      "other      146\n",
      "Name: gender, dtype: int64\n",
      "===> age\n",
      "===> employment\n",
      "employment    3569\n",
      "education     1888\n",
      "retired       1064\n",
      "unemployed     207\n",
      "Name: employment, dtype: int64\n",
      "===> occupation\n",
      "none         3159\n",
      "unskilled    3060\n",
      "skilled       463\n",
      "manager        46\n",
      "Name: occupation, dtype: int64\n",
      "===> main_activity\n",
      "work         3338\n",
      "education    1721\n",
      "None          998\n",
      "leisure       263\n",
      "health        256\n",
      "shopping      152\n",
      "Name: main_activity, dtype: int64\n",
      "===> main_activity_id\n",
      "===> main_facility_area\n",
      "===> main_facility_zone\n",
      "===> mode\n",
      "car      5510\n",
      "PT        144\n",
      "cycle      72\n",
      "walk        4\n",
      "Name: mode, dtype: int64\n",
      "===> hh_income_bin\n",
      "low         1792\n",
      "mid         1685\n",
      "high        1626\n",
      "mid-high    1625\n",
      "Name: hh_income_bin, dtype: int64\n",
      "===> age_bin\n",
      "21-31    1438\n",
      "51-61    1277\n",
      "4-13     1216\n",
      "61-71     697\n",
      "16-21     646\n",
      "31-41     612\n",
      "41-51     400\n",
      "0-4       280\n",
      "13-16      91\n",
      "71-81      71\n",
      "81+         0\n",
      "Name: age_bin, dtype: int64\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [],
   "outputs": [],
   "metadata": {}
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "9a55370b5416f60967d4532f4c3c966dd70cd0de3ce01b06e2ca05408dd0646e"
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.7.3 64-bit ('3.7.3': pyenv)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "358.4px"
   },
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}