{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Loading  Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/mark/GoogleDrive/UM/S4/GEMS/Git/SEMB_env/bin/python3\n",
      "3.6.2 (v3.6.2:5fd33b5926, Jul 16 2017, 20:11:06) \n",
      "[GCC 4.2.1 (Apple Inc. build 5666) (dot 3)]\n",
      "sys.version_info(major=3, minor=6, micro=2, releaselevel='final', serial=0)\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "print(sys.executable)\n",
    "print(sys.version)\n",
    "print(sys.version_info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Synthetic_Regular\n",
      "BlogCatalog\n",
      "ICEWS\n",
      "Facebook\n",
      "Synthetic_Role\n",
      "DD6\n",
      "PPI\n",
      "airports\n"
     ]
    }
   ],
   "source": [
    "from semb.datasets import load as load_dataset\n",
    "from semb.datasets import get_dataset_ids\n",
    "for did in get_dataset_ids():\n",
    "    print(did)\n",
    "    load_dataset(did)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These are the `dataset_id` for the existing datasets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 Load Specific Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Built-in Dataset (Graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get airports datasets\n",
    "DataProvider = load_dataset(\"airports\")\n",
    "Datasets = DataProvider().get_datasets()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`Datasets` contain all the datasets with the identifier `airports`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[DatasetInfo(name='0 - brazil', description='Brazil airport sample data', src_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_Brazil/brazil-airports.edgelist', label_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_Brazil/airport_Brazil_label.txt'),\n",
       " DatasetInfo(name='1 - european', description='European airport sample data', src_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_European/europe-airports.edgelist', label_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_European/airport_European_label.txt'),\n",
       " DatasetInfo(name='2 - US', description='European airport sample data', src_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_US/usa-airports.edgelist', label_url='/Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/datasets/airports/../../../sample-data/Airports/airport_US/airport_USA_label.txt')]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For example, to get the Br-Air traffic dataset, which is the first in the Datasets list. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_graph = DataProvider().load_dataset(Datasets[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The loaded graph will have the type of `networkx.Graph()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<networkx.classes.graph.Graph at 0x7f9648fb0f60>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset_graph"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Built-in Dataset (Graph)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the built-in dataset contains the corresponding label, call the `load_label()` function to get the label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Read in 131 node labels.\n",
      ">>> Label 0 appears 32 times\n",
      ">>> Label 1 appears 32 times\n",
      ">>> Label 3 appears 35 times\n",
      ">>> Label 2 appears 32 times\n"
     ]
    }
   ],
   "source": [
    "dataset_label = DataProvider().load_label(Datasets[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 Load Customized Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.1 Load Customized Dataset (Graph)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SEMB is compatible with all the *undirected, unweighted* `networkx.Graph()` data as input"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.2 Load Customized Dataset (Label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `get_label(input_dir, delimeter)` function to load customized label file. Please make sure each line of the label file aligns with the following format.\n",
    "\n",
    "\\<Node_id(int)\\> \\<delimeter\\> \\<Label_id(int)\\>\n",
    "\n",
    "The default delimeter is blank."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from semb.evaluations.utils import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Read in 131 node labels.\n",
      ">>> Label 2 appears 31 times\n",
      ">>> Label 1 appears 81 times\n",
      ">>> Label 0 appears 19 times\n"
     ]
    }
   ],
   "source": [
    "dict_customized_labels = \\\n",
    "get_label(\"./sample-data/labels/airport_Brazil_log_label.txt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Load Embedding Methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "graphwave\n",
      "degree2\n",
      "drne\n",
      "node2vec\n",
      "degree\n",
      "gae\n",
      "role2vec\n",
      "line\n",
      "degree1\n",
      "struc2vec\n",
      "xnetmf\n",
      "multilens\n",
      "segk\n",
      "riwalk\n"
     ]
    }
   ],
   "source": [
    "from semb.methods import load as load_method\n",
    "from semb.methods import get_method_ids\n",
    "for mid in get_method_ids():\n",
    "    print(mid)\n",
    "    load_method(mid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These are the `method_id` for the existing datasets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Load Specific Embedding Method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "EmbMethodClassXnetmf = load_method(\"xnetmf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Show the tunable parameters of the xNetMF method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dim': 128, 'max_layer': 2, 'discount': 0.1, 'gamma': 1}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EmbMethodClassXnetmf.__PARAMS__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Get Embedding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Embedding with xNetMF"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Initialize the embedding method with i) the target graph to embed 2)the tunable parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "xnetmf = EmbMethodClassXnetmf(dataset_graph,\n",
    "                              dim = 128,\n",
    "                              max_layer = 2,\n",
    "                              discount = 0.1,\n",
    "                              gamma = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `train()` function to start the embedding process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max degree:  80\n",
      "got k hop neighbors in time:  0.04764890670776367\n",
      "got degree sequences in time:  0.012601137161254883\n",
      "computed representation in time:  0.008405923843383789\n"
     ]
    }
   ],
   "source": [
    "xnetmf.train()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `get_embeddings()` function to get the embedding file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_xnetmf_emb = xnetmf.get_embeddings()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2 Embedding with struc2vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dim': 128,\n",
       " 'walk_length': 80,\n",
       " 'num_walks': 10,\n",
       " 'window_size': 10,\n",
       " 'until_layer': None,\n",
       " 'iter': 5,\n",
       " 'workers': 1,\n",
       " 'weighted': False,\n",
       " 'directed': False,\n",
       " 'opt1': False,\n",
       " 'opt2': False,\n",
       " 'opt3': False}"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EmbMethodClass = load_method(\"struc2vec\")\n",
    "EmbMethodClass.__PARAMS__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rm /Users/mark/GoogleDrive/UM/S4/GEMS/Git/StrucEmbeddingLibrary/semb/methods/struc2vec/pickles/weights_distances-layer-*.pickle\n"
     ]
    }
   ],
   "source": [
    "struc2vec = EmbMethodClass(dataset_graph, \n",
    "                           num_walks=10, \n",
    "                           walk_length=80, \n",
    "                           window_size=10, \n",
    "                           dim=128, \n",
    "                           opt1=True, opt2=True, opt3=True, until_layer=2)\n",
    "struc2vec.train()\n",
    "dict_struc2vec_emb = struc2vec.get_embeddings()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.3 Embedding with node2vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dim': 128,\n",
       " 'walk_length': 80,\n",
       " 'num_walks': 10,\n",
       " 'window_size': 10,\n",
       " 'iter': 1,\n",
       " 'workers': 1,\n",
       " 'p': 1,\n",
       " 'q': 1,\n",
       " 'weighted': False,\n",
       " 'directed': False}"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EmbMethodClass = load_method(\"node2vec\")\n",
    "EmbMethodClass.__PARAMS__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "node2vec = EmbMethodClass(dataset_graph, p=1, q=4, iter=5)\n",
    "node2vec.train()\n",
    "dict_node2vec_emb = node2vec.get_embeddings()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Classification and Clustering Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from semb.evaluations.classification import *\n",
    "from semb.evaluations.clustering import *\n",
    "from semb.evaluations.utils import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Classification "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.1.1 Perform Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `perform_classification(dict_emb, dict_label)` function to perform classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_xnetmf_classification = \\\n",
    "perform_classification(dict_xnetmf_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'overall': {'accuracy': {'mean': 0.7026, 'std': 0.0771},\n",
       "  'f1_macro': {'mean': 0.6841, 'std': 0.0785},\n",
       "  'f1_micro': {'mean': 0.7026, 'std': 0.0771},\n",
       "  'auc_micro': {'mean': 0.9022, 'std': 0.0462},\n",
       "  'auc_macro': {'mean': 0.9098, 'std': 0.0466}},\n",
       " 'detailed': {0: {'accuracy': 0.6667,\n",
       "   'f1_macro': 0.6293,\n",
       "   'f1_micro': 0.6667,\n",
       "   'auc_micro': 0.8989,\n",
       "   'auc_macro': 0.896},\n",
       "  1: {'accuracy': 0.5769,\n",
       "   'f1_macro': 0.5673,\n",
       "   'f1_micro': 0.5769,\n",
       "   'auc_micro': 0.822,\n",
       "   'auc_macro': 0.8317},\n",
       "  2: {'accuracy': 0.8077,\n",
       "   'f1_macro': 0.7939,\n",
       "   'f1_micro': 0.8077,\n",
       "   'auc_micro': 0.9591,\n",
       "   'auc_macro': 0.9641},\n",
       "  3: {'accuracy': 0.7308,\n",
       "   'f1_macro': 0.7049,\n",
       "   'f1_micro': 0.7308,\n",
       "   'auc_micro': 0.9334,\n",
       "   'auc_macro': 0.9499},\n",
       "  4: {'accuracy': 0.7308,\n",
       "   'f1_macro': 0.725,\n",
       "   'f1_micro': 0.7308,\n",
       "   'auc_micro': 0.8974,\n",
       "   'auc_macro': 0.9073}}}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_xnetmf_classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_struc2vec_classification = \\\n",
    "perform_classification(dict_struc2vec_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_node2vec_classification = \\\n",
    "perform_classification(dict_node2vec_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.1.2 Show Results in a Concatenated Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "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>methods</th>\n",
       "      <th>accuracy_mean</th>\n",
       "      <th>accuracy_std</th>\n",
       "      <th>f1_macro_mean</th>\n",
       "      <th>f1_macro_std</th>\n",
       "      <th>f1_micro_mean</th>\n",
       "      <th>f1_micro_std</th>\n",
       "      <th>auc_micro_mean</th>\n",
       "      <th>auc_micro_std</th>\n",
       "      <th>auc_macro_mean</th>\n",
       "      <th>auc_macro_std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>xnetmf</td>\n",
       "      <td>0.7026</td>\n",
       "      <td>0.0771</td>\n",
       "      <td>0.6841</td>\n",
       "      <td>0.0785</td>\n",
       "      <td>0.7026</td>\n",
       "      <td>0.0771</td>\n",
       "      <td>0.9022</td>\n",
       "      <td>0.0462</td>\n",
       "      <td>0.9098</td>\n",
       "      <td>0.0466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>struc2vec</td>\n",
       "      <td>0.7860</td>\n",
       "      <td>0.0676</td>\n",
       "      <td>0.7809</td>\n",
       "      <td>0.0675</td>\n",
       "      <td>0.7860</td>\n",
       "      <td>0.0676</td>\n",
       "      <td>0.9193</td>\n",
       "      <td>0.0261</td>\n",
       "      <td>0.9318</td>\n",
       "      <td>0.0148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>node2vec</td>\n",
       "      <td>0.3587</td>\n",
       "      <td>0.0706</td>\n",
       "      <td>0.3229</td>\n",
       "      <td>0.0644</td>\n",
       "      <td>0.3587</td>\n",
       "      <td>0.0706</td>\n",
       "      <td>0.5931</td>\n",
       "      <td>0.0316</td>\n",
       "      <td>0.5678</td>\n",
       "      <td>0.0338</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     methods  accuracy_mean  accuracy_std  f1_macro_mean  f1_macro_std  \\\n",
       "0     xnetmf         0.7026        0.0771         0.6841        0.0785   \n",
       "1  struc2vec         0.7860        0.0676         0.7809        0.0675   \n",
       "2   node2vec         0.3587        0.0706         0.3229        0.0644   \n",
       "\n",
       "   f1_micro_mean  f1_micro_std  auc_micro_mean  auc_micro_std  auc_macro_mean  \\\n",
       "0         0.7026        0.0771          0.9022         0.0462          0.9098   \n",
       "1         0.7860        0.0676          0.9193         0.0261          0.9318   \n",
       "2         0.3587        0.0706          0.5931         0.0316          0.5678   \n",
       "\n",
       "   auc_macro_std  \n",
       "0         0.0466  \n",
       "1         0.0148  \n",
       "2         0.0338  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "concatenate_result_pd([(\"xnetmf\", dict_xnetmf_classification), \n",
    "                       (\"struc2vec\", dict_struc2vec_classification),\n",
    "                       (\"node2vec\", dict_node2vec_classification)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.1.3 Bar-plot Visualization for Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from semb.evaluations.visualization import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2400x750 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_classification([(\"xnetmf\", dict_xnetmf_classification), \n",
    "                          (\"struc2vec\", dict_struc2vec_classification),\n",
    "                          (\"node2vec\", dict_node2vec_classification)],\n",
    "                         metric='f1_macro', error_bar=True, rotation_deg=45 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2.1 Perform Clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `perform_classification(dict_emb, dict_label)` function to perform classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/mark/GoogleDrive/UM/S4/GEMS/Git/SEMB_env/lib/python3.6/site-packages/sklearn/metrics/cluster/supervised.py:859: FutureWarning: The behavior of NMI will change in version 0.22. To match the behavior of 'v_measure_score', NMI will use average_method='arithmetic' by default.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "dict_xnetmf_clustering = \\\n",
    "perform_clustering(dict_xnetmf_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'overall': {'purity': 0.45038167938931295, 'nmi': 0.24750668135395176}}"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_xnetmf_clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/mark/GoogleDrive/UM/S4/GEMS/Git/SEMB_env/lib/python3.6/site-packages/sklearn/metrics/cluster/supervised.py:859: FutureWarning: The behavior of NMI will change in version 0.22. To match the behavior of 'v_measure_score', NMI will use average_method='arithmetic' by default.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "dict_struc2vec_clustering = \\\n",
    "perform_clustering(dict_struc2vec_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/mark/GoogleDrive/UM/S4/GEMS/Git/SEMB_env/lib/python3.6/site-packages/sklearn/metrics/cluster/supervised.py:859: FutureWarning: The behavior of NMI will change in version 0.22. To match the behavior of 'v_measure_score', NMI will use average_method='arithmetic' by default.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "dict_node2vec_clustering = \\\n",
    "perform_clustering(dict_node2vec_emb, dataset_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.2.2 Show Results in a Concatenated Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "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>methods</th>\n",
       "      <th>purity</th>\n",
       "      <th>nmi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>xnetmf</td>\n",
       "      <td>0.450382</td>\n",
       "      <td>0.247507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>struc2vec</td>\n",
       "      <td>0.656489</td>\n",
       "      <td>0.458403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>node2vec</td>\n",
       "      <td>0.343511</td>\n",
       "      <td>0.074570</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     methods    purity       nmi\n",
       "0     xnetmf  0.450382  0.247507\n",
       "1  struc2vec  0.656489  0.458403\n",
       "2   node2vec  0.343511  0.074570"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "concatenate_result_pd([(\"xnetmf\", dict_xnetmf_clustering), \n",
    "                       (\"struc2vec\", dict_struc2vec_clustering),\n",
    "                       (\"node2vec\", dict_node2vec_clustering)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.2.3 Bar-plot Visualization for Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2400x750 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_clustering([(\"xnetmf\", dict_xnetmf_clustering), \n",
    "                      (\"struc2vec\", dict_struc2vec_clustering),\n",
    "                      (\"node2vec\", dict_node2vec_clustering)],\n",
    "                    metric='purity', rotation_deg=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Perform Centrality Correlation Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "from semb.evaluations.centrality_correlation import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9420771391553502"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "centrality_correlation(dataset_graph, \n",
    "                       dict_struc2vec_emb, \n",
    "                       centrality='clustering_coeff', \n",
    "                       similarity='euclidean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3938822195894267"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "centrality_correlation(dataset_graph, \n",
    "                       dict_node2vec_emb, \n",
    "                       centrality='pagerank', \n",
    "                       similarity='cosine')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8457912846571526"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "centrality_correlation(dataset_graph, \n",
    "                       dict_xnetmf_emb, \n",
    "                       centrality='degree', \n",
    "                       similarity='euclidean')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "SEMB_env",
   "language": "python",
   "name": "semb_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}