{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---------------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# $SPA$tial $G$rap$H$s: n$ET$works, $T$opology, & $I$nference\n",
    "\n",
    "## Tutorial for `pysal.spaghetti`: Working with point patterns: empirical observations\n",
    "#### James D. Gaboardi [<jgaboardi@fsu.edu>]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Instantiating a `pysal.spaghetti.Network`\n",
    "2. Allocating observations to a network\n",
    "    * snapping\n",
    "3. Visualizing original and snapped locations\n",
    "    * visualization with `geopandas` and `matplotlib`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:25.919391Z",
     "start_time": "2019-05-14T00:25:25.896788Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "last_modified = None\n",
    "if os.name == \"posix\":\n",
    "    last_modified = !stat -f\\\n",
    "                    \"# This notebook was last updated: %Sm\"\\\n",
    "                     Spaghetti_Pointpatterns_Empirical.ipynb\n",
    "elif os.name == \"nt\":\n",
    "    last_modified = !for %a in (Spaghetti_Pointpatterns_Empirical.ipynb)\\\n",
    "                    do echo # This notebook was last updated: %~ta\n",
    "    \n",
    "if last_modified:\n",
    "    get_ipython().set_next_input(last_modified[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:25.933253Z",
     "start_time": "2019-05-14T00:25:25.926333Z"
    }
   },
   "outputs": [],
   "source": [
    "# This notebook was last updated: Dec  9 14:23:58 2018"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-----------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.016615Z",
     "start_time": "2019-05-14T00:25:25.942023Z"
    }
   },
   "outputs": [],
   "source": [
    "from pysal.explore import spaghetti as spgh\n",
    "from pysal.lib import examples\n",
    "import geopandas as gpd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.lines as mlines\n",
    "from shapely.geometry import Point, LineString\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "__author__ = \"James Gaboardi <jgaboardi@gmail.com>\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Instantiating a `pysal.spaghetti.Network`\n",
    "### Instantiate the network from `.shp` file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.155155Z",
     "start_time": "2019-05-14T00:25:27.026125Z"
    }
   },
   "outputs": [],
   "source": [
    "ntw = spgh.Network(in_data=examples.get_path('streets.shp'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Allocating observations to a network\n",
    "### Snap point patterns to the network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.473692Z",
     "start_time": "2019-05-14T00:25:27.167257Z"
    }
   },
   "outputs": [],
   "source": [
    "# Crimes with attributes\n",
    "ntw.snapobservations(examples.get_path('crimes.shp'),\n",
    "                     'crimes',\n",
    "                     attribute=True)\n",
    "\n",
    "# Schools without attributes\n",
    "ntw.snapobservations(examples.get_path('schools.shp'),\n",
    "                     'schools',\n",
    "                     attribute=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 3. Visualizing original and snapped locations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## True and snapped school locations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.494084Z",
     "start_time": "2019-05-14T00:25:27.475176Z"
    }
   },
   "outputs": [],
   "source": [
    "schools_df = spgh.element_as_gdf(ntw,\n",
    "                                 pp_name='schools',\n",
    "                                 snapped=False)\n",
    "\n",
    "snapped_schools_df = spgh.element_as_gdf(ntw,\n",
    "                                         pp_name='schools',\n",
    "                                         snapped=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## True and snapped crime locations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.522298Z",
     "start_time": "2019-05-14T00:25:27.498845Z"
    }
   },
   "outputs": [],
   "source": [
    "crimes_df = spgh.element_as_gdf(ntw,\n",
    "                                pp_name='crimes',\n",
    "                                snapped=False)\n",
    "\n",
    "snapped_crimes_df = spgh.element_as_gdf(ntw,\n",
    "                                        pp_name='crimes',\n",
    "                                        snapped=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create `geopandas.GeoDataFrame` objects of the vertices and arcs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.949426Z",
     "start_time": "2019-05-14T00:25:27.525105Z"
    }
   },
   "outputs": [],
   "source": [
    "# network nodes and edges\n",
    "vertices_df, arcs_df = spgh.element_as_gdf(ntw,\n",
    "                                           vertices=True,\n",
    "                                           arcs=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting `geopandas.GeoDataFrame` objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:27.964697Z",
     "start_time": "2019-05-14T00:25:27.952322Z"
    }
   },
   "outputs": [],
   "source": [
    "# legend patches\n",
    "arcs = mlines.Line2D([], [], color='k', label='Network Arcs', alpha=.5)\n",
    "vtxs = mlines.Line2D([], [], color='k', linewidth=0, markersize=2.5,\n",
    "                     marker='o', label='Network Vertices', alpha=1)\n",
    "schl = mlines.Line2D([], [], color='k', linewidth=0, markersize=25,\n",
    "                     marker='X', label='School Locations', alpha=1)\n",
    "snp_schl = mlines.Line2D([], [], color='k', linewidth=0, markersize=12,\n",
    "                         marker='o', label='Snapped Schools', alpha=1)\n",
    "crme = mlines.Line2D([], [], color='r', linewidth=0, markersize=7,\n",
    "                     marker='x', label='Crime Locations', alpha=.75)\n",
    "snp_crme = mlines.Line2D([], [], color='r', linewidth=0, markersize=3,\n",
    "                         marker='o', label='Snapped Crimes', alpha=.75)\n",
    "\n",
    "patches = [arcs, vtxs, schl, snp_schl, crme, snp_crme]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-05-14T00:25:28.589923Z",
     "start_time": "2019-05-14T00:25:27.969251Z"
    }
   },
   "outputs": [],
   "source": [
    "# plot figure\n",
    "base = arcs_df.plot(color='k', alpha=.25, figsize=(12,12), zorder=0)\n",
    "vertices_df.plot(ax=base, color='k', markersize=5, alpha=1)\n",
    "\n",
    "crimes_df.plot(ax=base, color='r', marker='x',\n",
    "               markersize=50, alpha=.5, zorder=1)\n",
    "snapped_crimes_df.plot(ax=base, color='r',\n",
    "                       markersize=20, alpha=.5, zorder=1)\n",
    "\n",
    "schools_df.plot(ax=base, cmap='tab20', column='id', marker='X',\n",
    "                markersize=500, alpha=.5, zorder=2)\n",
    "snapped_schools_df.plot(ax=base,cmap='tab20', column='id',\n",
    "                        markersize=200, alpha=.5, zorder=2)\n",
    "\n",
    "# add legend\n",
    "plt.legend(handles=patches, fancybox=True, framealpha=0.8,\n",
    "           scatterpoints=1, fontsize=\"xx-large\", bbox_to_anchor=(1.04, .6))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-----------"
   ]
  }
 ],
 "metadata": {
  "_draft": {
   "nbviewer_url": "https://gist.github.com/cbf4e5b556b77f3a03ff2e65c3481b63"
  },
  "anaconda-cloud": {},
  "gist": {
   "data": {
    "description": "notebooks/Spaghetti_Pointpatterns_Empirical.ipynb",
    "public": true
   },
   "id": "cbf4e5b556b77f3a03ff2e65c3481b63"
  },
  "kernelspec": {
   "display_name": "Python [conda env:py3_spgh_dev]",
   "language": "python",
   "name": "conda-env-py3_spgh_dev-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}