{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "from pprint import pprint\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import networkx as nx\n",
    "import seaborn as sns\n",
    "\n",
    "from graphrole import RecursiveFeatureExtractor, RoleExtractor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the well known karate_club_graph from Networkx\n",
    "G = nx.karate_club_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Features extracted from 3 recursive generations:\n",
      "    external_edges(mean)(mean)  degree(mean)  degree(sum)  \\\n",
      "0                    19.637500      4.312500         69.0   \n",
      "1                    22.422685      5.777778         52.0   \n",
      "2                    25.537083      6.600000         66.0   \n",
      "3                    23.717361      7.666667         46.0   \n",
      "4                    17.979167      7.666667         23.0   \n",
      "5                    17.234375      6.250000         25.0   \n",
      "6                    17.234375      6.250000         25.0   \n",
      "7                    26.342708     10.250000         41.0   \n",
      "8                    27.214363     11.800000         59.0   \n",
      "9                    28.108824     13.500000         27.0   \n",
      "10                   17.979167      7.666667         23.0   \n",
      "11                   24.937500     16.000000         16.0   \n",
      "12                   25.302083     11.000000         22.0   \n",
      "13                   26.897696     11.600000         58.0   \n",
      "14                   29.017157     14.500000         29.0   \n",
      "15                   29.017157     14.500000         29.0   \n",
      "16                   13.000000      4.000000          8.0   \n",
      "17                   26.302083     12.500000         25.0   \n",
      "18                   29.017157     14.500000         29.0   \n",
      "19                   27.240605     14.000000         42.0   \n",
      "20                   29.017157     14.500000         29.0   \n",
      "21                   26.302083     12.500000         25.0   \n",
      "22                   29.017157     14.500000         29.0   \n",
      "23                   25.340196      8.000000         40.0   \n",
      "24                   21.527778      4.333333         13.0   \n",
      "25                   21.477778      4.666667         14.0   \n",
      "26                   25.183824     10.500000         21.0   \n",
      "27                   25.871078      8.750000         35.0   \n",
      "28                   24.461438     11.000000         33.0   \n",
      "29                   24.908578      9.000000         36.0   \n",
      "30                   27.675245     10.750000         43.0   \n",
      "31                   27.773080      9.000000         54.0   \n",
      "32                   22.394526      5.083333         61.0   \n",
      "33                   22.418627      3.823529         65.0   \n",
      "\n",
      "    external_edges(mean)  degree  external_edges  internal_edges  \n",
      "0              24.937500      16              17              34  \n",
      "1              27.666667       9              19              21  \n",
      "2              27.100000      10              34              21  \n",
      "3              25.666667       6              20              16  \n",
      "4              16.000000       3              16               5  \n",
      "5              13.000000       4              15               7  \n",
      "6              13.000000       4              15               7  \n",
      "7              22.500000       4              25              10  \n",
      "8              25.000000       5              44              10  \n",
      "9              26.000000       2              25               2  \n",
      "10             16.000000       3              16               5  \n",
      "11             17.000000       1              15               1  \n",
      "12             18.500000       2              18               3  \n",
      "13             21.600000       5              41              11  \n",
      "14             20.500000       2              25               3  \n",
      "15             20.500000       2              25               3  \n",
      "16             15.000000       2               4               3  \n",
      "17             18.000000       2              21               3  \n",
      "18             20.500000       2              25               3  \n",
      "19             18.000000       3              37               4  \n",
      "20             20.500000       2              25               3  \n",
      "21             18.000000       2              21               3  \n",
      "22             20.500000       2              25               3  \n",
      "23             20.600000       5              27               9  \n",
      "24             26.666667       3               8               4  \n",
      "25             25.666667       3               9               4  \n",
      "26             21.000000       2              17               3  \n",
      "27             21.750000       4              29               5  \n",
      "28             31.333333       3              28               4  \n",
      "29             21.250000       4              24               8  \n",
      "30             26.000000       4              33               7  \n",
      "31             17.166667       6              42               9  \n",
      "32             28.916667      12              23              25  \n",
      "33             29.117647      17              18              32  \n"
     ]
    }
   ],
   "source": [
    "# extract features\n",
    "feature_extractor = RecursiveFeatureExtractor(G)\n",
    "features = feature_extractor.extract_features()\n",
    "\n",
    "print(f'\\nFeatures extracted from {feature_extractor.generation_count} recursive generations:')\n",
    "print(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Node role assignments:\n",
      "{0: 'role_3',\n",
      " 1: 'role_6',\n",
      " 2: 'role_6',\n",
      " 3: 'role_6',\n",
      " 4: 'role_1',\n",
      " 5: 'role_0',\n",
      " 6: 'role_0',\n",
      " 7: 'role_1',\n",
      " 8: 'role_2',\n",
      " 9: 'role_1',\n",
      " 10: 'role_1',\n",
      " 11: 'role_1',\n",
      " 12: 'role_1',\n",
      " 13: 'role_2',\n",
      " 14: 'role_1',\n",
      " 15: 'role_1',\n",
      " 16: 'role_4',\n",
      " 17: 'role_1',\n",
      " 18: 'role_1',\n",
      " 19: 'role_1',\n",
      " 20: 'role_1',\n",
      " 21: 'role_1',\n",
      " 22: 'role_1',\n",
      " 23: 'role_2',\n",
      " 24: 'role_4',\n",
      " 25: 'role_4',\n",
      " 26: 'role_1',\n",
      " 27: 'role_2',\n",
      " 28: 'role_1',\n",
      " 29: 'role_1',\n",
      " 30: 'role_1',\n",
      " 31: 'role_2',\n",
      " 32: 'role_6',\n",
      " 33: 'role_3'}\n",
      "\n",
      "Node role membership by percentage:\n",
      "    role_0  role_1  role_2  role_3  role_4  role_5  role_6\n",
      "0     0.10    0.02    0.02    0.42    0.02    0.02    0.42\n",
      "1     0.12    0.02    0.02    0.02    0.02    0.28    0.52\n",
      "2     0.09    0.02    0.22    0.02    0.02    0.22    0.41\n",
      "3     0.12    0.02    0.02    0.02    0.02    0.28    0.52\n",
      "4     0.17    0.40    0.03    0.17    0.03    0.03    0.17\n",
      "5     0.27    0.27    0.05    0.27    0.05    0.05    0.05\n",
      "6     0.27    0.27    0.05    0.27    0.05    0.05    0.05\n",
      "7     0.12    0.29    0.02    0.12    0.02    0.29    0.12\n",
      "8     0.02    0.02    0.47    0.02    0.11    0.11    0.25\n",
      "9     0.17    0.40    0.03    0.03    0.03    0.17    0.17\n",
      "10    0.17    0.40    0.03    0.17    0.03    0.03    0.17\n",
      "11    0.04    0.55    0.04    0.04    0.04    0.24    0.04\n",
      "12    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "13    0.02    0.02    0.45    0.02    0.02    0.24    0.24\n",
      "14    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "15    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "16    0.06    0.06    0.06    0.06    0.34    0.06    0.34\n",
      "17    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "18    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "19    0.03    0.32    0.32    0.14    0.03    0.14    0.03\n",
      "20    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "21    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "22    0.03    0.42    0.03    0.03    0.03    0.42    0.03\n",
      "23    0.02    0.11    0.25    0.02    0.11    0.25    0.25\n",
      "24    0.04    0.20    0.04    0.04    0.46    0.04    0.20\n",
      "25    0.04    0.20    0.04    0.04    0.46    0.04    0.20\n",
      "26    0.03    0.40    0.03    0.03    0.17    0.17    0.17\n",
      "27    0.15    0.15    0.35    0.03    0.15    0.03    0.15\n",
      "28    0.14    0.32    0.14    0.03    0.03    0.03    0.32\n",
      "29    0.14    0.32    0.03    0.03    0.03    0.32    0.14\n",
      "30    0.02    0.27    0.27    0.02    0.02    0.12    0.27\n",
      "31    0.11    0.02    0.47    0.02    0.25    0.02    0.11\n",
      "32    0.28    0.02    0.02    0.12    0.02    0.02    0.52\n",
      "33    0.10    0.02    0.02    0.42    0.02    0.02    0.42\n"
     ]
    }
   ],
   "source": [
    "# assign node roles\n",
    "role_extractor = RoleExtractor(n_roles=None)\n",
    "role_extractor.extract_role_factors(features)\n",
    "node_roles = role_extractor.roles\n",
    "\n",
    "print('\\nNode role assignments:')\n",
    "pprint(node_roles)\n",
    "\n",
    "print('\\nNode role membership by percentage:')\n",
    "print(role_extractor.role_percentage.round(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build color palette for plotting\n",
    "unique_roles = sorted(set(node_roles.values()))\n",
    "color_map = sns.color_palette('Paired', n_colors=len(unique_roles))\n",
    "# map roles to colors\n",
    "role_colors = {role: color_map[i] for i, role in enumerate(unique_roles)}\n",
    "# build list of colors for all nodes in G\n",
    "node_colors = [role_colors[node_roles[node]] for node in G.nodes]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot graph\n",
    "plt.figure()\n",
    "\n",
    "with warnings.catch_warnings():\n",
    "    # catch matplotlib deprecation warning\n",
    "    warnings.simplefilter('ignore')\n",
    "    nx.draw(\n",
    "        G,\n",
    "        pos=nx.spring_layout(G, seed=42),\n",
    "        with_labels=True,\n",
    "        node_color=node_colors,\n",
    "    )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}