Social network Graph Link Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p style=\"font-size:32px;text-align:center\"> <b>Social network Graph Link Prediction - Facebook Challenge</b> </p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problem statement: \n",
    "Given a directed social graph, have to predict missing links to recommend users (Link Prediction in graph)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Overview\n",
    "Taken data from facebook's recruting challenge on kaggle https://www.kaggle.com/c/FacebookRecruiting  \n",
    "data contains two columns source and destination eac edge in graph \n",
    "    - Data columns (total 2 columns):  \n",
    "    - source_node         int64  \n",
    "    - destination_node    int64  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mapping the problem into supervised learning problem:\n",
    "- Generated training samples of good and bad links from given directed graph and for each link got some features like no of followers, is he followed back, page rank, katz score, adar index, some svd fetures of adj matrix, some weight features etc. and trained ml model based on these features to predict link. \n",
    "- Some reference papers and videos :  \n",
    "    - https://www.cs.cornell.edu/home/kleinber/link-pred.pdf\n",
    "    - https://www3.nd.edu/~dial/publications/lichtenwalter2010new.pdf\n",
    "    - https://kaggle2.blob.core.windows.net/forum-message-attachments/2594/supervised_link_prediction.pdf\n",
    "    - https://www.youtube.com/watch?v=2M77Hgy17cg"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Business objectives and constraints:  \n",
    "- No low-latency requirement.\n",
    "- Probability of prediction is useful to recommend ighest probability links"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Performance metric for supervised learning:  \n",
    "- Both precision and recall is important so F1 score is good choice\n",
    "- Confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Importing Libraries\n",
    "# please do go through this python notebook: \n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import csv\n",
    "import pandas as pd#pandas to create small dataframes \n",
    "import datetime #Convert to unix time\n",
    "import time #Convert to unix time\n",
    "# if numpy is not installed already : pip3 install numpy\n",
    "import numpy as np#Do aritmetic operations on arrays\n",
    "# matplotlib: used to plot graphs\n",
    "import matplotlib\n",
    "import matplotlib.pylab as plt\n",
    "import seaborn as sns#Plots\n",
    "from matplotlib import rcParams#Size of plots  \n",
    "from sklearn.cluster import MiniBatchKMeans, KMeans#Clustering\n",
    "import math\n",
    "import pickle\n",
    "import os\n",
    "# to install xgboost: pip3 install xgboost\n",
    "import xgboost as xgb\n",
    "\n",
    "import warnings\n",
    "import networkx as nx\n",
    "import pdb\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: \n",
      "Type: DiGraph\n",
      "Number of nodes: 1862220\n",
      "Number of edges: 9437519\n",
      "Average in degree:   5.0679\n",
      "Average out degree:   5.0679\n"
     ]
    }
   ],
   "source": [
    "#reading graph\n",
    "if not os.path.isfile('data/after_eda/train_woheader.csv'):\n",
    "    traincsv = pd.read_csv('data/train.csv')\n",
    "    print(traincsv[traincsv.isna().any(1)])\n",
    "    print(traincsv.info())\n",
    "    print(\"Number of diplicate entries: \",sum(traincsv.duplicated()))\n",
    "    traincsv.to_csv('data/after_eda/train_woheader.csv',header=False,index=False)\n",
    "    print(\"saved the graph into file\")\n",
    "else:\n",
    "    g=nx.read_edgelist('data/after_eda/train_woheader.csv',delimiter=',',create_using=nx.DiGraph(),nodetype=int)\n",
    "    print(nx.info(g))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Displaying a sub graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: \n",
      "Type: DiGraph\n",
      "Number of nodes: 66\n",
      "Number of edges: 50\n",
      "Average in degree:   0.7576\n",
      "Average out degree:   0.7576\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if not os.path.isfile('train_woheader_sample.csv'):\n",
    "    pd.read_csv('data/train.csv', nrows=50).to_csv('train_woheader_sample.csv',header=False,index=False)\n",
    "    \n",
    "subgraph=nx.read_edgelist('train_woheader_sample.csv',delimiter=',',create_using=nx.DiGraph(),nodetype=int)\n",
    "# https://stackoverflow.com/questions/9402255/drawing-a-huge-graph-with-networkx-and-matplotlib\n",
    "\n",
    "pos=nx.spring_layout(subgraph)\n",
    "nx.draw(subgraph,pos,node_color='#A0CBE2',edge_color='#00bb5e',width=1,edge_cmap=plt.cm.Blues,with_labels=True)\n",
    "plt.savefig(\"graph_sample.pdf\")\n",
    "print(nx.info(subgraph))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of unique persons 1862220\n"
     ]
    }
   ],
   "source": [
    "# No of Unique persons \n",
    "print(\"The number of unique persons\",len(g.nodes()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 No of followers for each person"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "indegree_dist = list(dict(g.in_degree()).values())\n",
    "indegree_dist.sort()\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(indegree_dist)\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of Followers')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "indegree_dist = list(dict(g.in_degree()).values())\n",
    "indegree_dist.sort()\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(indegree_dist[0:1500000])\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of Followers')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.boxplot(indegree_dist)\n",
    "plt.ylabel('No Of Followers')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "90 percentile value is 12.0\n",
      "91 percentile value is 13.0\n",
      "92 percentile value is 14.0\n",
      "93 percentile value is 15.0\n",
      "94 percentile value is 17.0\n",
      "95 percentile value is 19.0\n",
      "96 percentile value is 21.0\n",
      "97 percentile value is 24.0\n",
      "98 percentile value is 29.0\n",
      "99 percentile value is 40.0\n",
      "100 percentile value is 552.0\n"
     ]
    }
   ],
   "source": [
    "### 90-100 percentile\n",
    "for i in range(0,11):\n",
    "    print(90+i,'percentile value is',np.percentile(indegree_dist,90+i))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "99% of data having followers of 40 only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.1 percentile value is 42.0\n",
      "99.2 percentile value is 44.0\n",
      "99.3 percentile value is 47.0\n",
      "99.4 percentile value is 50.0\n",
      "99.5 percentile value is 55.0\n",
      "99.6 percentile value is 61.0\n",
      "99.7 percentile value is 70.0\n",
      "99.8 percentile value is 84.0\n",
      "99.9 percentile value is 112.0\n",
      "100.0 percentile value is 552.0\n"
     ]
    }
   ],
   "source": [
    "### 99-100 percentile\n",
    "for i in range(10,110,10):\n",
    "    print(99+(i/100),'percentile value is',np.percentile(indegree_dist,99+(i/100)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 842.4x595.44 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "sns.set_style('ticks')\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(11.7, 8.27)\n",
    "sns.distplot(indegree_dist, color='#16A085')\n",
    "plt.xlabel('PDF of Indegree')\n",
    "sns.despine()\n",
    "#plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 No of people each person is following"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outdegree_dist = list(dict(g.out_degree()).values())\n",
    "outdegree_dist.sort()\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(outdegree_dist)\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of people each person is following')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlkAAAF1CAYAAADbfv+XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xl4lNXd//HPJCEsSSACyhbCIlJBLqqIVA1WWnkQ+5NHCcuEaMSCXqAohkUDYYuAoFKwlopAq1WipcMmxQcRLVKRPEiVR4sILoSALLImChNCSDL37w/M0IjDHZKcWTLv13X1Kpncuec7R4GP53ufcxyWZVkCAABAjYoIdAEAAAC1ESELAADAAEIWAACAAYQsAAAAAwhZAAAABhCyAAAADCBkAQAAGEDIAgAAMICQBQAAYAAhCwAAwICoQBcgSb/4xS/UqlWrQJcBAABg6+DBg9q6davtdUERslq1aqVVq1YFugwAAABbycnJlbqOdiEAAIABhCwAAAADCFkAAAAGELIAAAAMIGQBAAAYQMgCAAAwgJAFAABgACELAADAAEIWAACAAUZ2fF+1apXeeOMNSVJxcbF27dqlnJwcNWzY0MTbAQAABB0jISs5Odm75fyTTz6pAQMGELAAAEBYMXp24Weffabdu3dr2rRpJt8GAADUAu7iUn28N19WFX++SUy0uibE12hN1WE0ZC1atEijRo36ye+5XC65XC5JUkFBgckyAABACFiwcbcW/DO3yj/vcEj/yuyty+Pq1mBVVWcsZJ08eVJ79uzRjTfe+JPfdzqdcjqdkip/mjUAAKi9CotLFVs3StnDe1Tp5y9rEB00AUsyGLI++ugj3XzzzaZuDwAAaqGoSIeuS7ws0GXUCGNbOOTl5SkhIcHU7QEAQC1jSXIEuogaZGwm64EHHjB1awAAUAtZVX3iPUixGSkAAAgaDkftmcsiZAEAgKBgyapV7UJCFgAACAq0CwEAAAypRd1CQhYAAAgO5yayak/KImQBAICgQLsQAADAENqFAAAANY7VhQAAADWOdiEAAIAhtAsBAABqmGVJjlrUMCRkAQCAoGCpdvULCVkAACBo0C4EAACoYefahbUHIQsAAASF2tUsJGQBAIAg4qhF/UJCFgAACArskwUAAGAAqwsBAAAMqUXdQkIWAAAIEhYhCwAAoMZZYsd3AAAA2CBkAQCAoGBZFu1CAACAmnauXVh7ELIAAAAMIGQBAICgYFns+A4AAFDjaBcCAADAFiELAAAEBcuyatVUFiELAAAEBdqFAAAAsEXIAgAAwYHVhQAAADXPkkW7EAAAABcXZerGixYt0nvvvaeSkhINGTJEgwYNMvVWAACgFji3GWmgq6g5RkLW1q1b9cknn2jp0qUqKirSyy+/bOJtAABALXJuB4fak7KMhKzNmzerY8eOGjVqlNxut5544gkTbwMAABC0jISsgoICHTp0SAsXLtSBAwf00EMP6e23365VKwYAAAg3lmVp4MItyj3mNnL/wuJSXXl5rJF7B4KRkBUfH6/27dsrOjpa7du3V926dZWfn68mTZp4r3G5XHK5XJLOhTIAABDcPJa0bV+Bft46XtcmNDLyHjdd2dTIfQPBSMi6/vrrtWTJEv32t7/V0aNHVVRUpPj4+ArXOJ1OOZ1OSVJycrKJMgAAQA2yLEuS1PvqK/TobVcFuJrgZyRk/epXv9JHH32kgQMHyrIsTZ06VZGRkSbeCgAA+IkV6AJCjLEtHHjYHQCA2olHrCuHzUgBAECl/NAtZCFbJRGyAABApVg0DC8JIQsAAMAAQhYAAKiU8+3CwNYRKghZAAAABhCyAADAJalN5wuaRMgCAACVQrvw0hCyAABApZSvLiRjVQ4hCwAAwABCFgAAqBTahZeGkAUAACqlfCtSHnyvHEIWAACAAYQsAABQKdYP/ULahZVDyAIAAJXCyYWXhpAFAABgACELAABUyvnVhfQLK4OQBQAAKqc8ZAW2ipBByAIAADAgyu6C1atXV/yBqCg1b95c3bt3N1YUAAAIPt5jdZjKqhTbkLV27VoVFRXpuuuu0/bt21VcXKyoqCh17txZmZmZ/qgRAAAEAYt24SWxDVmlpaVasmSJIiIi5PF49OCDD+qll15SSkqKP+oDAAAISbbPZH333XcqLS2VdC5wff/995Kks2fPmq0MAAAEFe+xOvQLK8V2Jis1NVX9+vXTVVddpT179uiBBx7QwoULdcstt/ijPgAAECTY8f3S2IasQYMGqXfv3vrmm2+UmJioyy67TGVlZYqMjPRHfQAAACHJNmTt2rVLLpdLxcXF3tdmz55ttCgAABB8vO3CgFYROmxD1oQJE3TvvfeqefPm/qgHAAAEKev8Q1kBrSNU2Iaspk2batCgQf6oBQAAoNawDVmtWrXS4sWL1alTJ+9qgp49exovDAAABBfvZqQBriNU2IaskpIS5eXlKS8vz/saIQsAgDDkPSA6sGWECp8hq7S0VFFRUXryySf9WQ8AAECt4DNkZWRkaO7cuerbt6+3TWhZlhwOhzZs2OC3AgEAQHA4v7qQqazK8Bmy5s6d6/3/6667zm8FAQCA4GTRLrwktsfqrFy5UsnJyZoxY4a2bt0qj8fjj7oAAECQsbxzWagM2wffZ86cKUn6+OOPNWfOHO3bt08ffvih8cIAAEBwYiKrcmxD1quvvqotW7YoPz9f3bp106OPPuqPugAAQJChXXhpbEPWpk2bdOrUKfXp00c9e/bU1VdfXakb33333YqLi5MkJSQkcBQPAAAhjmbhpbENWS+99JKKi4v14Ycf6qmnnlJeXp42b9580Z8pP+cwOzu7ZqoEAABBg9WFlWMbst555x29//772rlzp7p06aIHH3zQ9qZffPGFioqKNGzYMJWWlmrs2LG69tpra6RgAAAgvZKTp70nTvv1Pd3Fped+QcaqFNuQ9fHHH6t///566qmnKn3TevXqafjw4Ro0aJD27t2rBx98UG+//baios6/ncvlksvlkiQVFBRUoXQAAMLT2VKPst7cqeioCNWLst0ooEY1ja2rDlfE+vU9Q5VtyPrtb3+r2bNna9q0aWrbtq0mTpyohISEi/5Mu3bt1KZNGzkcDrVr107x8fE6duyYWrRo4b3G6XTK6XRKkpKTk6v5MQAACB+eH55AT+99lR7u1SHA1cAX2/g7ZcoU3XXXXVq6dKn69++vSZMm2d50xYoVevrppyVJR44ckdvt1uWXX179agEAAEKEbcgqLi7WbbfdpoYNG6p3794qKyuzvenAgQN16tQpDRkyRGPGjNGsWbMqtAoBAED18QB6cLNNPmVlZfryyy/1s5/9TF9++WWlbhodHe09lgcAANQsi70UQoJtyJo8ebIyMzN19OhRNWvWTDNmzPBHXQAAwAabggY325DVuXNnrVy50h+1AACASig/Q5CMFdx8hqyePXv6/CG7zUgBAIA5tAtDg8+QRZACACC40S4Mbj5D1tixY+Xw8U+Ph9oBAAic8oksVhcGN58hKyUlxZ91AACASrLoF4YEnyGrR48ekiS3260XXnhBubm5atu2rR5++GG/FQcAAHyjXRjcbDcjzczMVMuWLTVmzBi1atVKEyZM8EddAADAB+axQoPtFg4FBQVKS0uTJHXq1Enr1683XhQAAPCNbmFoqNSxOseOHZMkHT9+XB6Px3hRAADAnq8FaggOtjNZ6enpSklJUVxcnNxuNzu+AwAQaD/MZBGxgpvPkPXcc89pzJgxOn36tDZs2KD8/Hw1btzYn7UBAICfYPFUVkjwGbI2bNigK664QtnZ2Tpx4kSF7zmdTuOFAQCAi6NbGNx8hqxZs2YpJydHZ8+e9T6TBQAAAs+iXRgSfIasrl27qmvXrkpKSlLXrl39WRMAALgImoWhwXZ1IQELAIDgUr7jO6sLg5ttyAIAAMGJjBXcKhWy3G63CgsLtXr1an3//femawIAABdBuzA02O6T9cQTTygpKUmffPKJPB6P3n33Xb3wwgv+qA0AAPwEHnwPDbYzWQcPHtRdd92l3NxcTZ8+XW632x91AQAAhDTbkFVSUqK33npLHTp0UH5+vr777jt/1AUAAHzwbkbKQ1lBzTZkPfDAA1q/fr1GjBih7Oxspaen+6MuAADgC+3CkODzmazS0lJFRUWpV69e6tWrlyTpoYce8lddAAAAIc1nyMrIyNDcuXPVt29f7z4clmXJ4XBow4YNfisQAABUVL66kG5hcPMZsubOnStJeu+99/xWDAAAsHd+dSEpK5ixGSkAAIABhCwAAEJM+epC2oXBjZAFAECIYTPS0GC74/vChQv15z//WfXq1fO+tnnzZqNFAQAAhDrbkLVu3Tp98MEHql+/vj/qAQAANlhdGBps24WtWrWqMIsFAAACy/qhX8jqwuBmO5NVUlKifv36qWPHjpIkh8Ph3d4BAAAAP802ZD344IP+qAMAAFSS5e0XBrQM2LBtF3bu3FkbN27Un//8Z/3jH//wzmgBAIDAImMFN9uQlZmZqZYtW2rMmDFq1aqVJkyYUKkbnzhxQrfeeqtyc3OrXSQAAECosW0XFhQUKC0tTZLUqVMnrV+/3vamJSUlmjp1Kg/MAwBggHefLJYXBjXbkFVcXKxjx47p8ssv1/Hjx+XxeGxv+swzzyglJUWLFy+ukSIBAPA3y7L0xeFTKi61/3vP3w5/fybQJaASbEPWY489ppSUFMXFxcntdmvGjBkXvX7VqlVq3LixbrnlFkIWACBkbfr6uIa+/K9Al3FRMdGRgS4BF2EbspKSkrRhwwbl5+erUaNGioy8+D/QlStXyuFwaMuWLdq1a5cyMjL04osv6vLLL69wncvlksvlknSuJQkAQDA5WVQiSXqqfxe1bBR8G3JHR0XoF+0aB7oMXESldnz3eDw6e/as5syZo+HDh2v48OE+r3/99de9v05LS1NWVtYFAUuSnE6nnE6nJCk5ObkqtQMAYEz5Lgm/aNdEHa6IDWgtCE22qwtffvll3XzzzVqzZo3++c9/auPGjf6oCwCAgPLuqs6z5agi25ms6OhoSVJMTIyio6NVWFhY6ZtnZ2dXvTIAAIAQZjuT1bp1aw0YMEADBgzQH//4R3Xt2tUfdQEAEBSYyEJV2c5k3X333ZoyZYpiYmLUpUuXn3y+CgCA2oa9qFBdtjNZ8+fPV0xMjCQRsAAAYcPyPvoOVI3tTJbD4dCoUaPUrl07RUScy2Rjx441XhgAAMGAeSxUlW3IGjBggD/qAAAgqJxvFwa2DoQu23Zhv379VFpaqv3796tly5a69dZb/VEXAAABZdEtRDXZhqxp06bp0KFDysnJUWFhoTIyMvxRFwAAQcFBwxBVZBuyvvnmGz322GOKjo7Wr3/9a506dcofdQEAEFDlE1m0C1FVtiGrrKxM+fn5cjgccrvd3offAQCozSz6hagm2wff09PTNWTIEB07dkxOp1OZmZn+qAsAACCk2YasHj16aO3atTp69KhatGjBpmwAgLBAuxDVZdv7e+edd9SnTx+NGjVKffr0UU5Ojj/qAgAgsOgWoppsZ7IWLFig5cuXq0mTJjp+/LhGjhyppKQkf9QGAEDA0cFBVdnOZMXHx6tJkyaSpKZNmyo2NtZ4UQAABFr5sTpELFSV7UxWbGyshg8frhtuuEGff/65zpw5o3nz5knieB0AQO3F4kJUl23Iuu2227y/btasmdFiAAAINnQLUVW2Iat///7+qAMAgKDiXV1IwxBVxM6iAAD8BNqFqC5CFgAAF0G7EFVl2y787LPP9MYbb6ioqMj72uzZs40WBQBAoLG6ENVlG7KysrJ07733qmnTpv6oBwCAoEC7ENVVqS0cePgdABC2mMpCFfkMWZs3b5YkxcXFaeHChbrmmmu8u9727NnTP9UBABAgrC5EdfkMWWvXrpV0LmTt27dP+/bt836PkAUAqPXoF6KafIas8ofb8/PztWvXLiUlJem1117Tf//3f/utOAAAAo3Vhagq2y0cxo0bp1OnTkmSGjVqpMcff9x4UQAABNr5diFQNbYhq6ioSH379pUk9evXT6dPnzZeFAAAgUa3ENVlG7Lq1KmjnJwcud1ubdmyRZGRkf6oCwCAoOCgX4gqsg1ZM2fO1Ouvv65Bgwbpr3/9q6ZPn+6PugAACCjLYjNSVI/tPllt2rTRggULvF8fPXrUaEEAAAQD7zNZpCxUkW3Iev7557V06VKVlJTozJkzatu2rXd7BwAAAPw023bhBx98oE2bNqlfv35666231KxZM3/UBQBAQJU/+M5mpKgq25AVHx+v6OhoFRYWqk2bNhUOigYAoLbyLi4kY6GKbENW8+bNtWLFCtWvX19z586V2+32R10AAAAhzfaZrOnTp+vw4cPq27ev3njjDT333HO2Ny0rK9PkyZOVl5enyMhIzZ49W4mJiTVSMAAA/uBdXchMFqrINmSdPn1aLpdLx44dU69evVSnTh3bm27cuFGS9Le//U1bt27V7Nmz9eKLL1a/WgAA/IyMhaqybRdmZmaqdevW2rt3r5o2bapJkybZ3rR3796aMWOGJOnQoUNq2rRp9SsFAAAIIbYzWd99950GDhyoNWvWqFu3bt7pU9sbR0UpIyND7777rv7whz9Uu1AAQPA5W+pR2ktbdfRUcaBLqXHfF5VIYsd3VJ1tyJKk3NxcSdLhw4cVEWE7+eX1zDPPaPz48Ro8eLDWrl2rBg0aeL/ncrnkcrkkSQUFBZdSMwAgSBScPqutefn6eUIjtWkSE+hyalzrxvUVE81xcqga25A1efJkZWZmKjc3V6NHj9a0adNsb7p69WodOXJEI0aMUP369eVwOC4489DpdMrpdEqSkpOTq1g+ACCQypsbKT0SNaQHC5yA/2Qbsjp27OidcaqsPn36aOLEibrnnntUWlqqzMxM1a1bt8pFAgCCk6XKPUIChKNKtQsvVYMGDfT888+buDUAIAjx1BJwoco/YAUAwI94j54hZQEXqNRM1pYtW7R//3517dpV7dq1o/UHAJAkmoXARdiGrHnz5unw4cPKzc1VnTp1tHjxYs2bN88ftQEAQgSHKAMXsm0Xbtu2Tc8++6waNGig/v3768CBA/6oCwAQArx7J5KxgAvYhqyysjIVFxfL4XCorKzskvbJAgDUbpXcnxoIS7btwqFDhyo5OVn5+fkaNGiQ7r//fj+UBQAIJUxkAReyDVl33HGHbr75Zu3bt08JCQlq3LixP+oCAIQQjp4BLuQzZI0dO9bnb5q5c+caKwgAEDpoFwK++QxZKSkpFb52OByVPhwaABBemMcCLuQzZPXo0UOSdOLECb344ovau3evrrrqKo0cOdJvxQEAglv5sTp0C4EL2S4VTE9P15VXXqnx48crISFBTzzxhD/qAgCEABocgG+V2vF9yJAhkqSrr75ab7/9ttGCAAChh5ks4EK2M1nt27fXmjVrdOTIEb333nuKj49XXl6e8vLy/FEfACCIlU9kseM7cCHbmaw9e/Zoz549WrFihffB96lTp8rhcGjJkiXGCwQABK/yvxeYyQIuZBuysrOzVVBQoP3797NPFgAAQCXZtgvXrVunlJQULVy4UE6nU3//+9/9URcAIATw3Dvgm+1M1iuvvKJVq1YpJiZGbrdbQ4cO1V133eWP2gAAQc57PjT9QuACtjNZDodDMTExkqTY2FjVrVvXeFEAAAChznYmKzExUU8//bS6d++ujz/+WImJif6oCwAQEn548D3AVQDByHYma9asWWrdurW2bNmi1q1ba8aMGf6oCwAQAs63CwNbBxCMKtUujI6OVqNGjdShQwedOnXKH3UBAACENNuQNXXqVB06dEg5OTkqLCxURkaGP+oCAIQANiMFfLMNWd98840ee+wxRUdH69e//jUzWQAAL9qFgG+2IausrEz5+flyOBxyu92KiLD9EQAAgLBnu7owPT1dQ4YM0bFjx+R0OpWZmemPugAAIcBidSHgk23I6tGjh9auXaujR4+qRYsWbDgHAPCiXQj4Ztv7e+edd9SnTx+NGjVKffr0UU5Ojj/qAgAACGm2M1kLFizQ8uXL1aRJEx0/flwjR45UUlKSP2oDAAQ5y3t4IVNZwI/ZzmTFx8erSZMmkqSmTZsqNjbWeFEAgNDgfSaLjAVcwHYmKzY2VsOHD9cNN9ygzz//XGfOnNG8efMkSWPHjjVeIAAAQCiyDVm33Xab99fNmjUzWgwAILR4H3wPbBlAULINWf379/dHHQCAEMbKc+BC7CwKAABggM+QtW7dOknSwYMH/VYMACC00C4EfPMZshYvXqyvv/5aY8aM0d69e5WXl+f9HwAAEqsLgYvx+UzWoEGDNGvWLOXl5WnKlCne1x0Oh5YsWeLzhiUlJcrMzNTBgwd19uxZPfTQQxUengcAAAgHPkNWamqqUlNTtWzZMg0ePLjSN1yzZo3i4+M1Z84cFRQUqH///oQsAKilOFYH8O2iqwvXr1+vN998U4sWLVLLli11zz33qG/fvhe9Yd++fXX77bd7v46MjKyZSgGghqz//LC++PZUoMuoFQ6fPCNJcvBUFnABnyFr9erVWrdunbKystS6dWvt3btXc+bMUWFhoQYMGODzhjExMZIkt9ut0aNHKz09/Sevc7lccrlckqSCgoLqfAYAuCQZK7fru9MlgS6j1qhXJ0It4+sHugwg6PgMWcuXL9df/vIXRUdHS5I6duyo3//+9xo+fPhFQ5Ykffvttxo1apRSU1PVr1+/n7zG6XTK6XRKkpKTk6taPwBcstIyS/ff3FZT7+wc6FJqjYgIZrKAH/MZsiIjI70Bq1xMTIxt++/48eMaNmyYpk6dqptuuqlmqgSAGhbhcBAMABjlcwsHj8ejwsLCCq+53W6VlZVd9IYLFy7UyZMntWDBAqWlpSktLU1nzpypmWoBoAZYlsWD2gCM8zmTdc899+iRRx7R+PHjlZiYqIMHD+rZZ5/Vvffee9EbTp48WZMnT67xQgGgplhi80wA5vkMWXfccYcaNmyo+fPna//+/WrevLnS0tL0q1/9yp/1AQAAhKSLbuGQlJSkpKQkf9UCAH5hWezrBMA8DogGEHYsWXKQsgAYRsgCAAAw4KLtwnLvv/++vv76a7Vt21a9e/c2XRMAGGVZPPgOwDzbmay5c+dqxYoVioqK0urVq/X000/7oy4AMMaSSFkAjLOdyfroo4/0t7/9TZI0dOjQSzosGgAAIFzZzmSVlpbK4/FIKt/Aj//8AxDiLA40BmCe7UzWb37zGw0ZMkQ///nPtX37dv3mN7/xR10AYMy51YWBrgJAbWcbsoYNG6aePXtqz549GjhwoDp27OiPugAAAEKaz5C1evXqC17buXOndu7cqbvvvttoUQBgEqsLAfiDz5CVm5tb4WvLsrRq1SrVq1ePkAUgpFlix3cA5vkMWePGjfP+et++fZowYYJ69eqlzMxMvxQGAAAQymyfyXr99df16quvauLEiRwODaBWsCyL1YUAjPMZso4cOaKJEyeqUaNGWr58uRo1auTPugDAGNqFAPzBZ8i68847VadOHd14442aPn16he/NnTvXeGEAAAChzGfIeuGFF/xZBwD4DasLAfiDz5DVo0cPf9YBAP5FvxCAYbbH6gAAAODSEbIAhBXLsiTRLgRgHiELQFj5IWPRLQRgHCELAADAAEIWgLDyw0QWm5ECMI6QBSCseJ/JImMBMIyQBSCsWPaXAECNIGQBCEtMZAEwjZAFIKywuhCAvxCyAIQVi4YhAD8hZAEISw6msgAYRsgCEFYsJrIA+AkhCwAAwABCFoCwRLcQgGmELABhxbu6kE0cABhGyAIQVlhdCMBfjIWsf//730pLSzN1ewCoFtqFAEyLMnHTP/3pT1qzZo3q169v4vYAUGXn24UAYJaRkJWYmKj58+friSeeMHF7hACPx9KBgqJAlwFc4HRJaaBLABAmjISs22+/XQcOHDBxa4SIZ9d/qYXv5wa6DMCnenUiA10CgFrOSMiqDJfLJZfLJUkqKCgIVBkw5Li7WPEN6mjK/+sc6FKAC0RFOnRbp2aBLgNALRewkOV0OuV0OiVJycnJgSoDhliWFBMdpQHXJwS6FAAAAoItHGAEy+QBAOHOWMhKSEjQsmXLTN0eAAAgqDGTBTMs9iECAIQ3QhaMsETIAgCEN0IWAACAAYQsGGFZFgfwAgDCGiELRtAuBACEO0IWAACAAYQsGGFZHMALAAhvhCwYca5dSMwCAIQvQhaMsCx2fAcAhDdCFoxhHgsAEM4IWTDCkkhZAICwRsiCGXQLAQBhjpAFY5jIAgCEM0IWjLBksboQABDWCFkwgsWFAIBwR8iCMcxjAQDCGSELRlgWZxcCAMIbIQtGWCwvBACEOUIWjHHQMAQAhDFCFoygXQgACHeELBhBsxAAEO4IWQAAAAYQsmDEuXYh/UIAQPgiZMEQGoYAgPBGyIIxzGMBAMIZIQtGsLoQABDuCFkwgmYhACDcEbJgDDNZAIBwRsiCEZZlseM7ACCsEbJgBO1CAEC4I2TBGNqFAIBwRsiCEZbFFg4AgPBGyIIRtAsBAOGOkAVz6BcCAMIYIQtGnFtdCABA+IoycVOPx6OsrCx9+eWXio6O1syZM9WmTRsTbwUAABCUjMxk/eMf/9DZs2flcrk0btw4Pf300ybeBkGObiEAIJwZmcnatm2bbrnlFknStddeqx07dph4m0rbkntCv3vnS3ksHsf2l91H3LqqWWygywAAIGCMhCy3263Y2PN/wUZGRqq0tFRRUeffzuVyyeVySZIKCgpMlOEVHRWh2LpRhCw/ujYxXnd0aRHoMgAACBgjISs2NlaFhYXerz0eT4WAJUlOp1NOp1OSlJycbKIMr+vbXKZXh/Uw+h4AAAD/ycgzWd26ddOmTZskSZ9++qk6duxo4m0AAACClpGZrP/6r/9STk6OUlJSZFmWZs2aZeJtAAAAgpaRkBUREaHp06ebuDUAAEBIYDNSAAAAAwhZAAAABhCyAAAADCBkAQAAGEDIAgAAMICQBQAAYAAhCwAAwABCFgAAgAGELAAAAAMIWQAAAAYYOVbnUh08eFDJyclG36OgoECXXXaZ0ffPE3bqAAAJsklEQVQIJYzHeYxFRYxHRYzHeYxFRYxHReE0HgcPHqzUdQ7LsizDtQSF5ORkrVq1KtBlBA3G4zzGoiLGoyLG4zzGoiLGoyLG40K0CwEAAAwgZAEAABgQmZWVlRXoIvylS5cugS4hqDAe5zEWFTEeFTEe5zEWFTEeFTEeFYXNM1kAAAD+RLsQAADAgFofsjwej6ZOnSqn06m0tDTt27cv0CXViJKSEj3++ONKTU3VwIEDtWHDBu3bt09DhgxRamqqpk2bJo/HI0n64x//qIEDByolJUXbt2+XpBq5NticOHFCt956q3Jzc8N+LBYtWiSn06nk5GQtX748rMejpKRE48aNU0pKilJTU8P2349///vfSktLk1Qzn6m61wbaf47Hrl27lJqaqrS0NA0fPlzHjx+XJC1btkzJyckaPHiwNm7cKEnKz8/XsGHDlJqaqvT0dBUVFdXItYH0n2NR7s0335TT6fR+HS5jUeOsWm79+vVWRkaGZVmW9cknn1gjR44McEU1Y8WKFdbMmTMty7Ks/Px869Zbb7VGjBhhffjhh5ZlWdaUKVOsd955x9qxY4eVlpZmeTwe6+DBg1ZycrJlWVa1rw02Z8+etR5++GGrT58+1u7du8N6LD788ENrxIgRVllZmeV2u60//OEPYT0e7777rjV69GjLsixr8+bN1iOPPBJ247F48WLrzjvvtAYNGmRZVvU/U01cG0g/Ho977rnH2rlzp2VZlrV06VJr1qxZ1tGjR60777zTKi4utk6ePOn99YwZM6yVK1dalmVZixYtsv7yl7/UyLWB8uOxsCzL2rlzp3Xfffd5XwuXsTCh1s9kbdu2Tbfccosk6dprr9WOHTsCXFHN6Nu3rx577DHv15GRkfr888/Vo0cPSdIvf/lL/e///q+2bdumnj17yuFwqGXLliorK1N+fn61rw02zzzzjFJSUnTFFVdIUliPxebNm9WxY0eNGjVKI0eOVK9evcJ6PNq1a6eysjJ5PB653W5FRUWF3XgkJiZq/vz53q/9+fl9XRtIPx6PefPmqVOnTpKksrIy1a1bV9u3b9d1112n6OhoxcXFKTExUV988UWFv1PKP2NNXBsoPx6LgoIC/e53v1NmZqb3tXAZCxNqfchyu92KjY31fh0ZGanS0tIAVlQzYmJiFBsbK7fbrdGjRys9PV2WZcnhcHi/f+rUqQs+f/nr1b02mKxatUqNGzf2/gaWFLZjIZ37Q3LHjh16/vnn9eSTT2r8+PFhPR4NGjTQwYMHdccdd2jKlClKS0sLu/G4/fbbFRV1/oAPf35+X9cG0o/Ho/w/zv7v//5Pr732mu6//3653W7FxcV5r4mJiZHb7a7w+n9+xupeGyj/ORZlZWWaNGmSMjMzFRMT470mXMbChKA4Vsek2NhYFRYWer/2eDwVfnOFsm+//VajRo1Samqq+vXrpzlz5ni/V1hYqIYNG17w+QsLCxUXF6eIiIhqXRtMVq5cKYfDoS1btmjXrl3KyMio8F/K4TQWkhQfH6/27dsrOjpa7du3V926dXX48GHv98NtPF555RX17NlT48aN07fffquhQ4eqpKTE+/1wGw9J1f5MNXFtsHnrrbf04osvavHixWrcuLHPustfr1evnu1nvJRrg8Hnn3+uffv2KSsrS8XFxdq9e7eeeuop3XjjjWE3FjWl1s9kdevWTZs2bZIkffrpp+rYsWOAK6oZx48f17Bhw/T4449r4MCBkqTOnTtr69atkqRNmzape/fu6tatmzZv3iyPx6NDhw7J4/GocePG1b42mLz++ut67bXXlJ2drU6dOumZZ57RL3/5y7AcC0m6/vrr9cEHH8iyLB05ckRFRUW66aabwnY8GjZs6P2Du1GjRiotLQ3b3yvl/Pn5fV0bTP7+9797/wxp3bq1JKlr167atm2biouLderUKeXm5qpjx47q1q2b3n//fUnnPuP1119fI9cGg65du2rt2rXKzs7WvHnz1KFDB02aNCksx6Km1Pp9sjwej7KysvTVV1/JsizNmjVLV155ZaDLqraZM2dq3bp1at++vfe1SZMmaebMmSopKVH79u01c+ZMRUZGav78+dq0aZM8Ho8mTpyo7t27Ky8vT1OmTKnWtcEoLS1NWVlZioiIqPbnC+WxePbZZ7V161ZZlqUxY8YoISEhbMejsLBQmZmZOnbsmEpKSnTfffepS5cuYTceBw4c0NixY7Vs2bIa+UzVvTbQysdj6dKluummm9SiRQvvzOMNN9yg0aNHa9myZXK5XLIsSyNGjNDtt9+u48ePKyMjQ4WFhbrssss0d+5cNWjQoNrXBsNYLFu2zOdr4TIWNa3WhywAAIBAqPXtQgAAgEAgZAEAABhAyAIAADCAkAUAAGAAIQsAAMAAQhaAoHHgwAENHjy40tcPHjxYBw4cqPJ7XXPNNRWO2lq6dGmFI0YAoDoIWQDCVmxsrCZOnKizZ88GuhQAtVDtOF8GQK2Tlpamq6++Wl9//bXcbreef/55tWrVSs8995w++OADNW/eXAUFBZKkU6dOadKkSd6vJ0+erIYNG2ro0KF67bXXlJubq/nz52vJkiUVjtVq06aNunfvrueee04ZGRkV3n/NmjV69dVXFR0drbZt22r69OmqU6eO/wYAQMhjJgtA0OratateeeUVJSUlae3atfrqq6/00UcfacWKFXr22We9554tXLhQN954o7KzszVjxgxlZWWpRYsWevzxxzVhwgTNnj1b8+bN+8lzS9PT05WTk6OPP/7Y+1pBQYHmz5+vV199VUuXLlVcXJxcLpffPjeA2oGQBSBode7cWZLUvHlz74G1Xbp0UUREhGJjY73nnH311VdauXKl0tLSNGXKFJ08eVKS1Lt3bx0+fFg9evRQ8+bNf/I9oqOjNXv2bE2ePFlFRUWSpP3796tDhw6KjY2VdO6Yla+//tr0xwVQyxCyAISMdu3aafv27fJ4PDp9+rR2794tSWrfvr3uv/9+ZWdn6/e//7369esnSXr55ZeVlJSkHTt26NNPP/V532uuuUZ33nmn/vSnP0mSEhISlJubq9OnT0uS/vWvf6ldu3aGPx2A2oZnsgCEjE6dOqlv374aOHCgrrjiCjVp0kSSNHLkSE2aNEnLli2T2+3WI488os8++0z/8z//I5fLpf379+vRRx+Vy+VSXFzcT9575MiR2rhxoySpcePGevTRR3XfffcpIiJCiYmJGj9+vN8+J4DagQOiAQAADKBdCAAAYAAhCwAAwABCFgAAgAGELAAAAAMIWQAAAAYQsgAAAAwgZAEAABhAyAIAADDg/wOC/sIKPmPtJgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "indegree_dist = list(dict(g.in_degree()).values())\n",
    "indegree_dist.sort()\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(outdegree_dist[0:1500000])\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of people each person is following')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.boxplot(indegree_dist)\n",
    "plt.ylabel('No Of people each person is following')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "90 percentile value is 12.0\n",
      "91 percentile value is 13.0\n",
      "92 percentile value is 14.0\n",
      "93 percentile value is 15.0\n",
      "94 percentile value is 17.0\n",
      "95 percentile value is 19.0\n",
      "96 percentile value is 21.0\n",
      "97 percentile value is 24.0\n",
      "98 percentile value is 29.0\n",
      "99 percentile value is 40.0\n",
      "100 percentile value is 1566.0\n"
     ]
    }
   ],
   "source": [
    "### 90-100 percentile\n",
    "for i in range(0,11):\n",
    "    print(90+i,'percentile value is',np.percentile(outdegree_dist,90+i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.1 percentile value is 42.0\n",
      "99.2 percentile value is 45.0\n",
      "99.3 percentile value is 48.0\n",
      "99.4 percentile value is 52.0\n",
      "99.5 percentile value is 56.0\n",
      "99.6 percentile value is 63.0\n",
      "99.7 percentile value is 73.0\n",
      "99.8 percentile value is 90.0\n",
      "99.9 percentile value is 123.0\n",
      "100.0 percentile value is 1566.0\n"
     ]
    }
   ],
   "source": [
    "### 99-100 percentile\n",
    "for i in range(10,110,10):\n",
    "    print(99+(i/100),'percentile value is',np.percentile(outdegree_dist,99+(i/100)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 842.4x595.44 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('ticks')\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(11.7, 8.27)\n",
    "sns.distplot(outdegree_dist, color='#16A085')\n",
    "plt.xlabel('PDF of Outdegree')\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of persons those are not following anyone are 274512 and % is 14.741115442858524\n"
     ]
    }
   ],
   "source": [
    "print('No of persons those are not following anyone are' ,sum(np.array(outdegree_dist)==0),'and % is',\n",
    "                                sum(np.array(outdegree_dist)==0)*100/len(outdegree_dist) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of persons having zero followers are 188043 and % is 10.097786512871734\n"
     ]
    }
   ],
   "source": [
    "print('No of persons having zero followers are' ,sum(np.array(indegree_dist)==0),'and % is',\n",
    "                                sum(np.array(indegree_dist)==0)*100/len(indegree_dist) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of persons those are not following anyone and also not having any followers are 0\n"
     ]
    }
   ],
   "source": [
    "count=0\n",
    "for i in g.nodes():\n",
    "    if len(list(g.predecessors(i)))==0 :\n",
    "        if len(list(g.successors(i)))==0:\n",
    "            count+=1\n",
    "print('No of persons those are not following anyone and also not having any followers are',count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.3 both followers + following "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "dict_in = dict(g.in_degree())\n",
    "dict_out = dict(g.out_degree())\n",
    "d = Counter(dict_in) + Counter(dict_out)\n",
    "in_out_degree = np.array(list(d.values()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAF1CAYAAABYusasAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xt0FPXh/vFncwVyIQREQIgEgQryjYIRqQnxhqIWvERgQzRCvZxCEQ2oDUYIVBC8JbY/EClWv2JASYBobfHWLxWQSwHjBUFRhBDuCCRAdsl1d35/ULZGiJsAmZ2w79c5HtnZyc4zc9bw+JmZz9gMwzAEAAAAywnwdQAAAACcHkUNAADAoihqAAAAFkVRAwAAsCiKGgAAgEVR1AAAACyKogYAAGBRFDUAAACLoqgBAABYFEUNAADAooJ8HeBcuPrqq3XRRRf5OgYAAIBXe/bs0bp16+q17nlR1C666CIVFBT4OgYAAIBXycnJ9V6XU58AAAAWRVEDAACwKIoaAACARVHUAAAALIqiBgAAYFGNVtS++uorpaWlSZIOHz6s0aNH65577lFKSop27twpScrPz1dycrKGDRumTz75RJJUUlKi+++/X6mpqUpPT1d5eXljRQQAALC0Rpme49VXX9V7772n5s2bS5JeeOEFDR48WLfddpv+/e9/a/v27WrevLlyc3O1ZMkSVVZWKjU1VQkJCZo9e7YGDRqk5ORkzZ07V3l5eRo5cmRjxAQAALC0RhlRi4mJ0cyZMz2vP//8cx04cEAjR47U3//+d/Xt21cbN25U7969FRISooiICMXExGjLli0qLCxU//79JUlJSUlas2ZNY0QEAACwvEYpagMHDlRQ0H8H6/bs2aPIyEi98cYbat++vV599VU5HA5FRER41gkLC5PD4ai1PCwsTGVlZY0REQAAwPJMuZkgKipKN9xwgyTphhtu0KZNmxQeHi6n0+lZx+l0KiIiotZyp9OpyMjI035mXl6ekpOTlZycrNLS0sbfCQAAAJOZUtSuvPJKrVixQpK0YcMGde3aVXFxcSosLFRlZaXKysq0bds2de/eXX369PGsu3LlSl155ZWn/Uy73a6CggIVFBSoVatWZuwGAACAqUx51mdGRoYmTpyohQsXKjw8XNnZ2WrZsqXS0tKUmpoqwzA0btw4hYaGavTo0crIyFB+fr5atWql7OxsMyICAAA/d/R4tSprXGob2czXUTxshmEYvg5xtpKTk3koOwAAOCsTlmzU9wfKVPD7hEbdTkN6CxPeAgAASCqrqNGxihpfx6iFogYAAGBRFDUAAACLoqgBAABYFEUNAADAoihqAAAAFkVRAwAAsCiKGgAAgEVR1AAAACyKogYAAGBRFDUAAACLoqgBAABYFEUNAADAoihqAAAAFkVRAwAAsCiKGgAAgEVR1AAAACyKogYAAGBRFDUAAACLoqgBAABYFEUNAADAoihqAAAAFkVRAwAAsCiKGgAAgEVR1AAAACyKogYAAGBRFDUAAACLoqgBAABYFEUNAADAoihqAAAAFtVoRe2rr75SWlparWV///vfZbfbPa/z8/OVnJysYcOG6ZNPPpEklZSU6P7771dqaqrS09NVXl7eWBEBAAAsrVGK2quvvqqJEyeqsrLSs+zbb7/V4sWLZRiGJOngwYPKzc3VwoUL9dprryknJ0dVVVWaPXu2Bg0apLfeeks9e/ZUXl5eY0QEAACwvEYpajExMZo5c6bndWlpqV588UVlZmZ6lm3cuFG9e/dWSEiIIiIiFBMToy1btqiwsFD9+/eXJCUlJWnNmjWNEREAAMDyghrjQwcOHKjdu3dLklwul5566illZmYqNDTUs47D4VBERITndVhYmBwOR63lYWFhKisrO+028vLyPKNtpaWljbEbAAAAPtUoRe2nNm/erOLiYk2ZMkWVlZX64Ycf9Mwzz6hfv35yOp2e9ZxOpyIiIhQeHi6n06lmzZrJ6XQqMjLytJ9rt9s917slJyc39m4AAACYrtHv+oyLi9PSpUuVm5urnJwcde3aVU899ZTi4uJUWFioyspKlZWVadu2berevbv69OmjFStWSJJWrlypK6+8srEjAgAAWFKjj6jV5YILLlBaWppSU1NlGIbGjRun0NBQjR49WhkZGcrPz1erVq2UnZ3tq4gAAAA+ZTNO3obZhCUnJ6ugoMDXMQAAQBM2ZsHn+u5Amf5v/LWNup2G9BYmvAUAALAor0Vt69at+uKLL/TVV19pxIgRWrt2rRm5AAAA/J7XojZ58mSFhITolVde0bhx4zRr1iwzcgEAAPg9r0UtKChI3bp1U3V1ta644gq5XC4zcgEAAPg9r0XNZrPpscceU1JSkt5//301b97cjFwAAAB+z+v0HC+99JK+/vprJSUlad26dXrppZfMyAUAAOD3vI6ojRkzRtdee61sNpv69eunqKgoM3IBAAD4Pa8jai1bttS8efMUGxurgIATvS4xMbHRgwEAAPg7r0WtVatW2rJli7Zs2eJZRlEDAABofF6L2owZM1RUVKSdO3fqV7/6ldq2bWtGLgAAAL/ntajNnz9f//znP3X06FHdddddKi4uVlZWlhnZAAAA/JrXmwmWLl2qN954QxERERoxYoS++uorM3IBAAD4Pa9F7eQz2202myQpJCSkcRMBAABAUj1Off7mN7/RPffco7179+qhhx7SgAEDzMgFAADg97wWtZSUFF1zzTX6/vvvFRsbq0svvdSMXAAAAH7Pa1Gz2+2KjY3VzTffrIsvvtiMTAAAAFA9ilpBQYG2bdumZcuW6be//a1at26tl19+2YxsAAAAfs1rUduyZYtWr16tdevWSZIuueSSRg8FAACAehS1e+65R506ddK4ceN07bXXmpEJAAAAqkdRW7dunQoLC7Vq1Sq9/vrrat26tXJycszIBgAA4Ne8zqN27Ngx7d+/X3v37lVFRYU6dOhgRi4AAAC/53VE7cEHH9SAAQM0evRode3a1YxMAAAAUD1G1PLz89WyZUstWLBA8+bNU1VVlRm5AAAA/J7XopaVlaVdu3YpISFBe/bs0cSJE83IBQAA4Pe8nvosLi7WggULJEkDBgxQSkpKo4cCAABAPUbUKisrVV5eLkmqqKiQy+Vq9FAAAACox4jafffdpzvuuEPdunXTDz/8oLFjx5qRCwAAwO95LWq33367kpKStGvXLnXs2FGtWrUyIxcAAIDfq7OojR8/Xjab7bTvZWdnN1ogAAAAnFBnUTvbmwa++uorvfjii8rNzdW3336rqVOnKjAwUCEhIXruuefUpk0b5efna+HChQoKCtLo0aN1/fXXq6SkRI8//rgqKirUtm1bzZgxQ82bNz+rLAAAAE1RnUWtqKiozh/q27fvL37oq6++qvfee89TsJ555hlNmjRJPXr00MKFC/Xqq6/qwQcfVG5urpYsWaLKykqlpqYqISFBs2fP1qBBg5ScnKy5c+cqLy9PI0eOPLO9AwAAaMLqvOvz4MGDdf7jTUxMjGbOnOl5nZOTox49ekiSXC6XQkNDtXHjRvXu3VshISGKiIhQTEyMtmzZosLCQvXv31+SlJSUpDVr1pztPgIAADRJdY6oPfzww54/L1++XFu3blVsbKwGDBjg9UMHDhyo3bt3e163bdtWkvT5559r/vz5WrBggT799FNFRER41gkLC5PD4ZDD4fAsDwsLU1lZWcP3CgAA4Dzg9a7P7OxsFRcXq0+fPnr33XdVWFiojIyMBm/o/fff1yuvvKK5c+cqOjpa4eHhcjqdnvedTqciIiI8y5s1ayan06nIyMjTfl5eXp7y8vIkSaWlpQ3OAwAAYHVeJ7zdsGGD/t//+38aOXKkZs6cqc8++6zBG/nb3/6m+fPnKzc3V506dZIkxcXFqbCwUJWVlSorK9O2bdvUvXt39enTRytWrJAkrVy5UldeeeVpP9Nut6ugoEAFBQVMGQIAAM5LXkfUampq5Ha7FRAQIMMw6pyyoy4ul0vPPPOM2rdv75ks96qrrtIjjzyitLQ0paamyjAMjRs3TqGhoRo9erQyMjKUn5+vVq1aMRUIAADwW16L2m233abhw4fr8ssv18aNG3XbbbfV64M7duyo/Px8SdL69etPu86wYcM0bNiwWsvatGmj1157rV7bAAAAOJ/VWdQ++OAD3XrrrRo4cKASExO1fft2DRkyRN27dzczHwAAgN+q8xq1uXPnauvWrRo3bpxCQkL0q1/9SsHBwb84vxoAAADOnTpH1IYOHarp06erqKhIWVlZMgxDkmSz2fTmm2+aFhAAAMBf1VnUUlNTlZqaqvz8/FOuIwMAAEDj8zo9ByUNAADAN7wWNQAAAPhGvYoaM/8DAACYr15F7dFHH23sHAAAAPgZr0Vt+/btqqioUFFREVNzAAAAmMjrkwkmT56sHTt2KCsri6k5AAAATOS1qOXm5iotLU25ublm5AEAAMB/1Osatd69ezd2DgAAAPxMvYra+PHjGzsHAAAAfoZ51AAAACyKogYAAGBRXm8mePfdd2v/QFCQ2rVrp/j4+EYLBQAAgHoUtaVLl6q8vFy9e/fWxo0bVVlZqaCgIPXs2VOZmZlmZAQAAPBLXotaTU2N3nzzTQUEBMjtduuhhx7Sa6+9ppSUFDPyAQAA+C2v16gdOXJENTU1kk6UtqNHj0qSqqqqGjcZAACAn/M6opaamqrBgwerW7du2r59ux588EHNmTNH/fv3NyMfAACA3/Ja1IYOHaoBAwZo586diomJUatWreRyuRQYGGhGPgAAAL/ltah9++23ysvLU2VlpWfZjBkzGjUUAAAA6lHUJkyYoHvvvVft2rUzIw8AAAD+w2tRa9OmjYYOHWpGFgAAAPyE16J20UUXae7cuerRo4dsNpskKTExsdGDAQAA+DuvRa26ulpFRUUqKiryLKOoAQCA840hw9cRTlFnUaupqVFQUJD++Mc/mpkHAADAZ2y+DvAzdRa1jIwMZWdn65ZbbvGc8jQMQzabTcuWLTMtIAAAgL+qs6hlZ2d7/t27d2/TAgEAAPiCYb0zn94fIbVkyRIlJydr6tSpWrdundxutxm5AAAATGez2LlPr0Vt2rRpKigo0K233qqcnBxdc8019frgr776SmlpaZKk4uJiDR8+XKmpqZo8ebKn7M2aNUtDhgxRSkqKNm7c+IvrAgAA+BuvRW3evHkaNWqUnn/+efXu3Vsvvvii1w999dVXNXHiRM/TDGbMmKH09HS99dZbMgxDy5Yt0+bNm7V+/XotWrRIOTk5npsWTrcuAACAP/Ja1FauXKmSkhLdfPPNuvPOO+s1NUdMTIxmzpzpeb1582b17dtXkpSUlKQ1a9aosLBQiYmJstls6tChg1wul0pKSk67LgAAQGOz4jVqXudRe+2111RZWal///vfeuaZZ1RUVKRVq1b94s8MHDhQu3fv9rw+ebeoJIWFhamsrEwOh0NRUVGedU4uP926p5OXl6e8vDxJUmlpqbfdAAAA8MpmsQk6vBa1jz/+WCtWrNA333yjXr166aGHHmrwRgIC/jtw53Q6FRkZqfDwcDmdzlrLIyIiTrvu6djtdtntdklScnJygzMBAABYnddTn5999pnuuusuvfPOO5o6daqSkpIavJGePXtq3bp1kk6cSo2Pj1efPn20atUqud1u7d27V263W9HR0addFwAAoLFZ8ckEXovab3/7W7355pv6zW9+ozFjxtQ6pVlfGRkZmjlzpux2u6qrqzVw4ED16tVL8fHxstvtGjt2rLKysupcFwAAwAxWm57DZhi/fOncgw8+qOHDh+uqq67S+vXrlZubq3nz5pmVr16Sk5NVUFDg6xgAAKAJ+13uZyo+fFwfpjf87GFDNKS3eB1Rq6ys1I033qjIyEgNGDBALpfrrAMCAADAO69FzeVy6bvvvpMkz78BAADON01yeo6JEycqMzNTP/74oy688EJNnTrVjFwAAAB+z2tR69mzp5YsWWJGFgAAAPxEnUXtl55A4G3CWwAAgKbGgmc+6y5qlDEAAOBvbBabn6POojZ+/Pg6w2ZnZzdaIAAAAJxQZ1FLSUkxMwcAAIBPNam7Pvv27StJcjgcevnll7Vt2zZ17txZv//9700LBwAAYCZrnfisxzxqmZmZ6tChg8aNG6eLLrpIEyZMMCMXAACA3/M6PUdpaanS0tIkST169NBHH33U6KEAAABQz0dIHTx4UJJ06NAhud3uRg8FAABgPutdpOZ1RC09PV0pKSmKiIiQw+HgyQQAAOC8ZbHZOeouai+99JLGjRun48ePa9myZSopKVF0dLSZ2QAAAPxanUVt2bJlatu2rXJzc3X48OFa79nt9kYPBgAAYKYmNT3H9OnTtXr1alVVVXmuUQMAADifNZlTn3FxcYqLi1NCQoLi4uLMzAQAAADV465PShoAAIBveC1qAAAA/sCCl6jVr6g5HA45nU69++67Onr0aGNnAgAA8AmbxR4i5XUetT/84Q9KSEjQF198IbfbrX/+8596+eWXzcgGAADg17yOqO3Zs0d33HGHtm3bpqeffloOh8OMXAAAAKYyLDg/h9eiVl1drffff19du3ZVSUmJjhw5YkYuAAAA01lteg6vRe3BBx/URx99pN/97nfKzc1Venq6GbkAAAD8Xp3XqNXU1CgoKEjXXXedrrvuOknS6NGjzcoFAADg9+osahkZGcrOztYtt9wi23/GAQ3DkM1m07Jly0wLCAAAYAbrXaH2C0UtOztbkvSvf/3LtDAAAAC+ZLFL1JjwFgAAwKooagAAAJIsODuH9wlvz5Xq6mpNmDBBe/bsUUBAgKZOnaqgoCBNmDBBNptN3bp10+TJkxUQEKBZs2Zp+fLlCgoKUmZmJs8bBQAA5rDY/Bxei9qcOXP017/+Vc2aNfMsW7VqVYM3tGLFCtXU1GjhwoVavXq1/vSnP6m6ulrp6em6+uqrlZWVpWXLlqlDhw5av369Fi1apH379mns2LFasmRJg7cHAADQ1Hktah988IE+/fRTNW/e/Kw2FBsbK5fLJbfbLYfDoaCgIH355Zfq27evJCkpKUmrV69WbGysEhMTZbPZ1KFDB7lcLpWUlCg6Ovqstg8AANDUeC1qF110Ua3RtDPVokUL7dmzR7feeqtKS0s1Z84cbdiwwTP1R1hYmMrKyuRwOBQVFeX5uZPLKWoAAKAxWfASNe9Frbq6WoMHD1b37t0lSTabzTN1R0O88cYbSkxM1GOPPaZ9+/ZpxIgRqq6u9rzvdDoVGRmp8PBwOZ3OWssjIiJO+by8vDzl5eVJkkpLSxucBwAA4OesdYVaPYraQw89dE42FBkZqeDgYElSy5YtVVNTo549e2rdunW6+uqrtXLlSvXr108xMTF64YUX9MADD2j//v1yu92nHU2z2+2y2+2SpOTk5HOSEQAAwEq8FrWePXvq5Zdf1rZt29S5c2f9/ve/P6MNjRw5UpmZmUpNTVV1dbXGjRunXr16adKkScrJyVGXLl00cOBABQYGKj4+Xna7XW63W1lZWWe0PQAAgIYwLDg/h83wkuqRRx7RVVddpfj4eK1fv15r167VnDlzzMpXL8nJySooKPB1DAAA0ISlvbZOjsoavfP7hEbdTkN6i9cRtdLSUqWlpUmSevTooY8++ujs0gEAAKBevD6ZoLKyUgcPHpQkHTp0SG63u9FDAQAAoB4jao8++qhSUlIUEREhh8OhqVOnmpELAADA73ktagkJCVq2bJlKSkrUsmVLBQYGmpELAADAdFabnsPrqc8PPvhAS5cu1YoVK9S/f3+99tprZuQCAADwe16L2uuvv65rrrlG7733npYvX65PPvnEjFwAAACmsuDsHN6LWkhIiKQTj3IKCQmp9dQAAACA88nJR1tahdei1qlTJ9199926++67NWvWLMXFxZmRCwAAwO95vZngzjvv1KRJkxQWFqZevXrpggsuMCMXAACA3/M6ojZz5kyFhYVJEiUNAACctwxZ7yI1ryNqNptNY8aMUWxsrAICTvS68ePHN3owAAAAs1nrCrV6FLW7777bjBwAAAD4Ga+nPgcPHqyamhrt2rVLHTp00LXXXmtGLgAAAFM1yek5Jk+erL1792r16tVyOp3KyMgwIxcAAIDpLDY7h/eitnPnTj366KMKCQnRDTfcoLKyMjNyAQAA+D2vRc3lcqmkpEQ2m00Oh8NzQwEAAAAal9ebCdLT0zV8+HAdPHhQdrtdmZmZZuQCAAAwlRWvUfNa1Pr27aulS5fqxx9/VPv27S33aAUAAIBzxWaxCTq8nsf8+OOPdfPNN2vMmDG6+eabtXr1ajNyAQAA+D2vI2qzZ8/WokWL1Lp1ax06dEijRo1SQkKCGdkAAABMY8UnE3gdUYuKilLr1q0lSW3atFF4eHijhwIAAPAJa5359D6iFh4ergceeEBXXXWVNm/erIqKCuXk5EjiUVIAAACNyWtRu/HGGz1/vvDCCxs1DAAAgK80ybs+77rrLjNyAAAA+JzFznx6v0YNAAAAvkFRAwAAsCivpz6//vprvfPOOyovL/csmzFjRqOGAgAAMJsh65369FrUpkyZonvvvVdt2rQxIw8AAIDPWO0BTPWanoMbCgAAAMxXZ1FbtWqVJCkiIkJz5szRZZdd5nnOZ2JiojnpAAAAzGLBc591FrWlS5dKOlHUiouLVVxc7HnvTIvaX/7yF/3rX/9SdXW1hg8frr59+2rChAmy2Wzq1q2bJk+erICAAM2aNUvLly9XUFCQMjMzFRcXd0bbAwAAaAirPZS9zqJ28oaBkpISffvtt0pISND8+fN1++23n9GG1q1bpy+++EJvv/22ysvL9frrr2vGjBlKT0/X1VdfraysLC1btkwdOnTQ+vXrtWjRIu3bt09jx47VkiVLzmzvAAAAmjCv03M89thjKisrkyS1bNlSTzzxxBltaNWqVerevbvGjBmjUaNG6brrrtPmzZvVt29fSVJSUpLWrFmjwsJCJSYmymazqUOHDnK5XCopKTmjbQIAADRlXm8mKC8v1y233CJJGjx4sPLz889oQ6Wlpdq7d6/mzJmj3bt3a/To0TIMw3PdW1hYmMrKyuRwOBQVFeX5uZPLo6Oja31eXl6e8vLyPJ8NAABwNgwLXqTmtagFBwdr9erVuvzyy/X1118rMDDwjDYUFRWlLl26KCQkRF26dFFoaKj279/ved/pdCoyMlLh4eFyOp21lkdERJzyeXa7XXa7XZKUnJx8RpkAAAB+ymrTc3g99Tlt2jQtWLBAQ4cO1VtvvaWnn376jDZ05ZVX6tNPP5VhGDpw4IDKy8v161//WuvWrZMkrVy5UvHx8erTp49WrVolt9utvXv3yu12nzKaBgAA4A+8jqhdfPHFmj17tuf1jz/+eEYbuv7667VhwwYNGTJEhmEoKytLHTt21KRJk5STk6MuXbpo4MCBCgwMVHx8vOx2u9xut7Kyss5oewAAAA1hGL5OcCqvRe3Pf/6z3n77bVVXV6uiokKdO3f2TN3RUH/4wx9OWTZ//vxTlo0dO1Zjx449o20AAACcqSZ36vPTTz/VypUrNXjwYL3//vu68MILzcgFAADg97wWtaioKIWEhMjpdOriiy+u9XB2AAAANB6vRa1du3ZavHixmjdvruzsbDkcDjNyAQAAmMqCl6h5v0bt6aef1v79+3XLLbfonXfe0UsvvWRGLgAAANM1mUdInXT8+HHl5eXp4MGDuu666xQcHGxGLgAAAL/n9dRnZmamOnXqpB07dqhNmzZ66qmnzMgFAABgKsOC83N4LWpHjhzRkCFDFBQUpD59+lhyJwAAAM6FJjc9hyRt27ZNkrR//34FBNTrRwAAAHCWvLauiRMnKjMzU998840eeeQRTZgwwYxcAAAAfs/rzQTdu3dXXl6eGVkAAAB8xooXd3EeEwAAwKIoagAAABbl9dSnJK1du1a7du1SXFycYmNjFRoa2ti5AAAATGXFiS28FrWcnBzt379f27ZtU3BwsObOnaucnBwzsgEAAJjKZrH5Obye+iwsLNTzzz+vFi1a6K677tLu3bvNyAUAAOD3vBY1l8ulyspK2Ww2uVwu5lEDAAAwiddTnyNGjFBycrJKSko0dOhQjRw50oRYAAAA5rLgJWrei9qtt96qa665RsXFxerYsaOio6PNyAUAAGA6a12h9gtFbfz48XVeUJednd1ogQAAAHBCnUUtJSWl1mubzcYD2QEAwPnLgj2nzqLWt29fSdLhw4f1yiuvaMeOHerWrZtGjRplWjgAAAAzWWx2Du93faanp+uSSy7R448/ro4dO+oPf/iDGbkAAAD8Xr2eTDB8+HBJ0qWXXqoPP/ywUQMBAADgBK8jal26dNF7772nAwcO6F//+peioqJUVFSkoqIiM/IBAACYwnpXqNVjRG379u3avn27Fi9e7LmZICsrSzabTW+++WajBwQAADCLxS5R817UcnNzVVpaql27djGPGgAAgIm8nvr84IMPlJKSojlz5shut+tvf/ubGbkAAABMZcHZObyPqL3xxhsqKChQWFiYHA6HRowYoTvuuMOMbAAAAKaqa7J/X/E6omaz2RQWFiZJCg8PV2hoaKOHAgAAQD1G1GJiYvTss88qPj5en332mWJiYszIBQAA4Pe8jqhNnz5dnTp10tq1a9WpUydNnTr1rDZ4+PBhXXvttdq2bZuKi4s1fPhwpaamavLkyXK73ZKkWbNmaciQIUpJSdHGjRvPansAAAD1YVhwgo56nfoMCQlRy5Yt1bVrV5WVlZ3xxqqrq5WVlaVmzZpJkmbMmKH09HS99dZbMgxDy5Yt0+bNm7V+/XotWrRIOTk5+uMf/3jG2wMAAGgIa12hVo+ilpWVpb1792r16tVyOp3KyMg4440999xzSklJUdu2bSVJmzdv9jxTNCkpSWvWrFFhYaESExNls9nUoUMHuVwulZSUnPE2AQAAmiqvRW3nzp169NFHFRISohtuuOGMR9QKCgoUHR2t/v37e5YZhuG5uyIsLExlZWVyOBwKDw/3rHNy+c/l5eUpOTlZycnJKi0tPaNMAAAAJzXJ6TlOjmjZbDY5HA4FBHjtdqe1ZMkS2Ww2rV2mAXM8AAAZY0lEQVS7Vt9++60yMjJqjZQ5nU5FRkYqPDxcTqez1vKIiIhTPs9ut8tut0uSkpOTzygTAADAT1lsdg7vI2rp6ekaPny4Nm3aJLvdrjFjxpzRhhYsWKD58+crNzdXPXr00HPPPaekpCStW7dOkrRy5UrFx8erT58+WrVqldxut/bu3Su3283TEAAAgF/yOqLWt29fLV26VD/++KPat29/TieCy8jI0KRJk5STk6MuXbpo4MCBCgwMVHx8vOx2u9xut7Kyss7Z9gAAAOrSJE99fvzxx3r22WfVsmVLORwOTZkyRQkJCWe10dzcXM+f58+ff8r7Y8eO1dixY89qGwAAAA1nrXOfXova7NmztWjRIrVu3VqHDh3SqFGjzrqoAQAAwDuv16hFRUWpdevWkqQ2bdrUuiMTAAAAjcfriFp4eLgeeOABXXXVVdq8ebMqKiqUk5MjSRo/fnyjBwQAADCDBS9R817UbrzxRs+fL7zwwkYNAwAA4EtWm57Da1G76667zMgBAACAnzmz2WsBAADOM4YF5+eos6h98MEHkqQ9e/aYFgYAAMCXLHbms+6iNnfuXG3dulXjxo3Tjh07VFRU5PkHAAAAja/Oa9SGDh2q6dOnq6ioSJMmTfIst9lsevPNN00JBwAA4M/qLGqpqalKTU1Vfn6+hg0bZmYmAAAAyMvNBB999JH+/ve/68Ybb1RaWpo+/PBDs3IBAACYrslMz/Huu+/qgw8+0JQpU9SpUyft2LFDL7zwgpxOp+6++24zMwIAAPilOkfUFi1apJkzZ+qSSy5RSEiIunfvrj/96U9atGiRmfkAAABMsae0XG6LzdBRZ1ELDAxUSEhIrWVhYWEKDAxs9FAAAABmu7BlMx1yVPo6Ri11FjW32y2n01lrmcPhkMvlavRQAAAAZnO7DXVs1cLXMWqps6jdc889evjhh7V582aVlZVpy5YteuSRR3TvvfeamQ8AAMAU1W63ggKsdTdBnTcT3HrrrYqMjNTMmTO1a9cutWvXTmlpabr++uvNzAcAAGAKl8toOkVNkhISEpSQkGBWFgAAAJ+pcRsKCrRWUeOh7AAAAJJcbkOBFhtRo6gBAABIOl7lUlCAtarRL576PGnFihXaunWrOnfurAEDBjR2JgAAANOVV7tUXmWt2S281sbs7GwtXrxYQUFBevfdd/Xss8+akQsAAMBUgQE2hQY3sRG1DRs2aOHChZKkESNG8IB2AABw3jEMQy63oagWId5XNpHX2lhTUyO32y3pxE7YrPa0UgAAgLNU5TrRdUKDmtiI2m233abhw4fr8ssv18aNG3XbbbeZkQsAAMA0VTUnilqwxabn8FrU7r//fiUmJmr79u0aMmSIunfvbkYuAAAA01T+p6g5Kmp8nKS2Oovau+++e8qyb775Rt98843uvPPORg0FAABgppMjah2imvs4SW11FrVt27bVem0YhgoKCtSsWTOKGgAAOK+cLGpN5q7Pxx57zPPn4uJiTZgwQdddd50yMzNNCQYAAGAWR+WJU54hgYE+TlKb12vUFixYoHnz5unJJ588qweyV1dXKzMzU3v27FFVVZVGjx6trl27asKECbLZbOrWrZsmT56sgIAAzZo1S8uXL1dQUJAyMzMVFxd3xtsFAADwpsZtSJKq/3P3p1XUWdQOHDigJ598Ui1bttSiRYvUsmXLs9rQe++9p6ioKL3wwgsqLS3VXXfdpUsvvVTp6em6+uqrlZWVpWXLlqlDhw5av369Fi1apH379mns2LFasmTJWW0bAADglxyvOjGidmFkMx8nqa3OojZo0CAFBwerX79+evrpp2u9l52d3eAN3XLLLRo4cKDndWBgoDZv3qy+fftKkpKSkrR69WrFxsYqMTFRNptNHTp0kMvlUklJiaKjoxu8TQAAgPo4WFYpqQlNz/Hyyy+f0w2FhYVJkhwOhx555BGlp6frueee80ygGxYWprKyMjkcDkVFRdX6ubKyMooaAABodFEtgn0doZY6i9rJka5zad++fRozZoxSU1M1ePBgvfDCC573nE6nIiMjFR4eLqfTWWt5RETEKZ+Vl5envLw8SVJpaek5zwoAAPzH7tJySVJYqNfL901l2j2ohw4d0v33368nnnhCQ4YMkST17NlT69atkyStXLlS8fHx6tOnj1atWiW32629e/fK7XafdjTNbreroKBABQUFatWqlVm7AQAAzkMhgScqUYtgaxU109LMmTNHx44d0+zZszV79mxJ0lNPPaVp06YpJydHXbp00cCBAxUYGKj4+HjZ7Xa53W5lZWWZFREAAPipPUdOjKiFN7NWUbMZhmH4OsTZSk5OVkFBga9jAACAJirrb5v05tpi7Xj2N42+rYb0FmtNvwsAAOADu0qO64KIUF/HOAVFDQAA+L3AgABVVLt8HeMUFDUAAOD3fvixTD3aRfo6xikoagAAwO+VV7tUWcOIGgAAgOU4KmrU1mKPj5IoagAAwM+53IacVS51aRPm6yinoKgBAAC/9mNZhSQp3GJPJZAoagAAwM8VHTzx6MqO0c19nORUFDUAAODXNu89Jknq1vbUZ4v7GkUNAAD4tU17j0qSurYN93GSU1HUAACAX/tuf5lsNqlZcKCvo5yCogYAAPyWYRjasr9MV8a08nWU06KoAQAAv3Xy+rReF7X0cZLTo6gBAAC/tXDDTknS7Vd08HGS06OoAQAAv7Vw/S5JUu9OUT5OcnoUNQAA4Jc++HqfatyGBl52oWw2m6/jnBZFDQAA+KWp//hGkvTH23v5OEndKGoAAMDvfLhpv/YerdBv/qe92rW03sPYT6KoAQAAv7Jx9xGNeetzdW7dQtnDLvd1nF9EUQMAAH5j1dZDuuPl1Qq02fTXEVdZcpLbn7LeY+IBAADOsRqXW0//4xu9ubZYIYEBKvj9NZZ8ZNTPUdQAAMB5y+02NG/tDj37wRZV1rjV/cJwvXn/1Za+Lu2nKGoAAOC8s/2gQ/+7eofeXr9TNW5DwYE2PX3HZbrv1519Ha1BKGoAAKDJq6px69/bD+vjb/Zr+XcHtbu0XJLUKbq57uvXWfcnxiowwJpzpf0SihoAAGhSqmrc2nHYqS92lurrPUf12Y5Sbdlf5nm/bUSoRl7TWUPjO+qyDtZ8hmd9UdQAAIAl7T9aoeLDTv1w0KHiw8e1/aBD3x9waGfJ8VrrdWjZTL+Ja6++naM18LJ2Teb6s/qgqAEAANNU1bjlNgztPVKu0uPVKj7s1NHyahUfPq7DzirtKT2u/UcrtPdoxSk/27J5sLpcEKa+sdHqfmG4enVoqT4Xt7L8FBtng6IGAAAa7HhVjRwVNZKkkuNV2vefYrW75LiOVdSoqsatXSXHVelyq9RZpf1HK1RR7dK+YxUyjNN/ZqsWwWoVFqLu7SJ07a/aql1kM8VeEKYubcLUuU2YwkP9r7b43x4DAOBnDMPQ9kPOWgXpyPGqWqNWFdUu7So5riqXW5L+W7RqTrw+VlGjvUfKZRiG3IZU4qzyut2gAJs6tmquwACbLoxspjYRoWoeHKDObcIUGhSoi6NbKLxZkC65IFzNQwL9soh5wxEBAOAsVbvccrnrGCbSiRK0/ZCzzvcPllXqYFnlKcsNSTsPO+Wscp3y3tHy6v8Up9rLfzxWoUM/K1FV/ylb9RESFKCT90Y2Cw5U59YtFBBgU4BN6tUhUh2imkuSggMD1Cm6hZoFByjQZlOn6BZqERKo4MAAdW0bLptNCgoIaJJ3WlqJJYua2+3WlClT9N133ykkJETTpk3TxRdf7OtYAIBfcNhRqV/oKl45KmtOuUjcG7fbUNEhp2fUpz6OlFdp75EKGXWdf/sPQ9KukuNyVNb84nqV1W7tO1p+VvvuTVhIoJqHnPpXdpvwELWNrH3hfMvmwbo2qplaNg/5xXVtkjq3DlPzkMBa60S1qP1z8C1LFrX/+7//U1VVlfLy8vTll1/q2Wef1SuvvOLrWMB55cjxKh2ux6kLXzrsqNKBY6deUGwFzsoa7S4tl8vLX/Zn68djlTrkOHWk5Vwp/cm1RWfjaHl1g0ZtfC0iNEgXRIZ6XS84IECxPyszp/Pr4Naekae6XBTVXC2bB5/2vQCbTV0uCFNI0KmP4A4JDKA8+TFLFrXCwkL1799fknTFFVdo06ZNPk4kPfvBFq0rOuzrGH6hrlMAOHcMNexUCOoWEnjqX6znlE1q37JZo/1FbbPZdEWnKLUJ915avGkdFqILz2JaBJuki1o1V2Sz05eZukQ2C1Kn6BYN+pnQoADZbJySg/VZsqg5HA6Fh//3QamBgYGqqalRUNB/4+bl5SkvL0+SVFpa2uiZwrjI0TQRzYKVcEkbRYU17Jc1GiYkMECdWrVQaHAjF42zEGCzKbZNmJpZNOOFkc0U0cBSAQANYcnmER4eLqfzvxddut3uWiVNkux2u+x2uyQpOTm50TONvbFbo28DAADgpyz5v6l9+vTRypUrJUlffvmlunfv7uNEAAAA5rPkiNpNN92k1atXKyUlRYZhaPr06b6OBAAAYDpLFrWAgAA9/fTTvo4BAADgU5Y89QkAAACKGgAAgGVR1AAAACyKogYAAGBRFDUAAACLoqgBAABYFEUNAADAoihqAAAAFkVRAwAAsCiKGgAAgEVZ8hFSDbVnzx4lJyc3+nZKS0vVqlWrRt9OU8HxqI3jURvHozaOR20cj9o4Hqc6n4/Jnj176r2uzTAMoxGznFeSk5NVUFDg6xiWwfGojeNRG8ejNo5HbRyP2jgep+KYnMCpTwAAAIuiqAEAAFhU4JQpU6b4OkRT0qtXL19HsBSOR20cj9o4HrVxPGrjeNTG8TgVx4Rr1AAAACyLU58AAAAWdV5Mz9GY3G63pkyZou+++04hISGaNm2aLr74Yl/HOieqq6uVmZmpPXv2qKqqSqNHj1a7du00atQode7cWZI0fPhw3XbbbZo1a5aWL1+uoKAgZWZmKi4uTsXFxZowYYJsNpu6deumyZMnKyAgoEHrWs2dd96piIgISVLHjh1lt9v1zDPPKDAwUImJiXr44Yfr/E58+eWXZ7Wu1RQUFOidd96RJFVWVurbb79Vdna2nn/+ebVv316SNHbsWMXHx5/3x+Orr77Siy++qNzc3HPyvT/bdX3tp8fj22+/1dSpUxUYGKiQkBA999xzatOmjaZNm6bPP/9cYWFhkqTZs2erurpajz/+uCoqKtS2bVvNmDFDzZs3V35+vhYuXKigoCCNHj1a119/vUpKSuq9rq/99Hhs3rzZ1N+hVv9+jBs3TocOHZJ0YkqKyy+/XC+99JJGjRqlI0eOKDg4WKGhofrrX/963h6Ps2bgF3300UdGRkaGYRiG8cUXXxijRo3ycaJzZ/Hixca0adMMwzCMkpIS49prrzXy8/ON1157rdZ6mzZtMtLS0gy3223s2bPHSE5ONgzDMH73u98Z//73vw3DMIxJkyYZH3/8cYPWtZqKigrjjjvuqLXs9ttvN4qLiw232208+OCDxqZNm+r8TpztulY2ZcoUY+HChUZOTo7x4Ycf1nrvfD8ec+fONQYNGmQMHTrUMIyz/96fi3V96efH45577jG++eYbwzAM4+233zamT59uGIZhpKSkGIcPH671s1OnTjWWLFliGIZh/OUvfzH+93//1/jxxx+NQYMGGZWVlcaxY8c8f27Iur708+Nh5u/QpvD9OOnIkSPG7bffbhw4cMAwDMO49dZbDbfbXWud8/F4nAvWG9KwmMLCQvXv31+SdMUVV2jTpk0+TnTu3HLLLXr00Uc9rwMDA7Vp0yYtX75c99xzjzIzM+VwOFRYWKjExETZbDZ16NBBLpdLJSUl2rx5s/r27StJSkpK0po1axq0rtVs2bJF5eXluv/++3Xfffdpw4YNqqqqUkxMjGw2mxITE7V27drTficcDsdZr2tVX3/9tX744QfZ7XZt3rxZS5YsUWpqqp599lnV1NSc98cjJiZGM2fO9Lw+2+/9uVjXl35+PHJyctSjRw9JksvlUmhoqNxut4qLi5WVlaWUlBQtXrxYUu3fpyf3cePGjerdu7dCQkIUERGhmJgYbdmypUHr+tLPj4eZv0ObwvfjpJkzZ+ree+9V27ZtdejQIR07dkyjRo3S8OHD9cknn0hqvP+2mjpOfXrhcDgUHh7ueR0YGKiamhoFBTX9Q3fylITD4dAjjzyi9PR0VVVVaejQoerVq5deeeUVvfzyy4qIiFBUVFStnysrK5NhGLLZbLWWORyOeq9rNc2aNdMDDzygoUOHaseOHXrooYcUGRnpeT8sLEy7du067Xfi58vOZF2r+stf/qIxY8ZIkhISEjRgwAB17NhRkydP1sKFC8/74zFw4EDt3r3b8/psv/fnYt3o6OhG3edf8vPj0bZtW0nS559/rvnz52vBggU6fvy47r33Xv32t7+Vy+XSfffdp169esnhcHguLfjpPp5cdnK5w+Fo0Lq+9PPjERcXZ9rv0Kbw/ZCkw4cPa+3atXryySclnbjs5uT/EB89elTDhw9XXFzceXk8zgVG1LwIDw+X0+n0vHa73edFSTtp3759uu+++3THHXdo8ODBuummmzy3Q99000365ptvTjkGTqdTERERta4xczqdioyMbNC6VhMbG6vbb79dNptNsbGxioiI0JEjRzzv17WPbrf7tPvd0HWt6NixY9q+fbv69esnSbr77rvVqVMn2Ww23Xjjjaf9fpzPx0PSWX/vz8W6VvP+++9r8uTJmjt3rqKjo9W8eXPdd999at68ucLDw9WvXz9t2bKl1v5428eGrGslZv4ObQrHQ5I+/PBDDRo0SIGBgZKkNm3aKCUlRUFBQWrdurV69OihoqIivzkeDUVR86JPnz5auXKlJOnLL79U9+7dfZzo3Dl06JDuv/9+PfHEExoyZIgk6YEHHtDGjRslSWvXrtVll12mPn36aNWqVXK73dq7d6/cbreio6PVs2dPrVu3TpK0cuVKxcfHN2hdq1m8eLGeffZZSdKBAwdUXl6uFi1aaOfOnTIMQ6tWrfLs48+/E+Hh4QoODj6rda1ow4YNuuaaaySdGEm6/fbbtX//fkm1vx/+cjwknfX3/lysayV/+9vfNH/+fOXm5qpTp06SpB07dig1NVUul0vV1dX6/PPPPd+VFStWSDqxj1deeaXi4uJUWFioyspKlZWVadu2berevXuD1rUSM3+HNoXvh3TiOCQlJXler1mzRunp6ZJOlKmtW7eqS5cufnM8Gur8GRpqJDfddJNWr16tlJQUGYah6dOn+zrSOTNnzhwdO3ZMs2fP1uzZsyVJEyZM0PTp0xUcHKw2bdpo6tSpCg8PV3x8vOx2u9xut7KysiRJGRkZmjRpknJyctSlSxcNHDhQgYGB9V7XaoYMGaInn3xSw4cPl81m0/Tp0xUQEKDHH39cLpdLiYmJuvzyy/U///M/p/1O/PGPfzyrda2oqKhIHTt2lCTZbDZNmzZNDz/8sJo1a6ZLLrlEw4YNU2BgoN8cD+nsv/fnYl2rcLlceuaZZ9S+fXuNHTtWknTVVVfpkUce0eDBgzVs2DAFBwfrjjvuULdu3TR69GhlZGQoPz9frVq1UnZ2tlq0aKG0tDSlpqbKMAyNGzdOoaGhDVrXSqZMmaKpU6ea8jvU6t+Pk4qKijwlXpKuvfZarVq1SsOGDVNAQIDGjx+v6OhovzkeDcWEtwAAABbFqU8AAACLoqgBAABYFEUNAADAoihqAAAAFkVRAwAAsCiKGoDzxu7duzVs2LB6rz9s2LBTZlFvyLYuu+yyWo+Ve/vtt0/7+BwAOFMUNQA4Q+Hh4XryySdVVVXl6ygAzlNMeAvgvJSWlqZLL71UW7dulcPh0J///GdddNFFeumll/Tpp5+qXbt2Ki0tlSSVlZXpqaee8ryeOHGiIiMjNWLECM2fP1/btm3TzJkz9eabb9Z6hNzFF1+s+Ph4vfTSS8rIyKi1/ffee0/z5s1TSEiIOnfurKefflrBwcHmHQAA5wVG1ACct+Li4vTGG28oISFBS5cu1ffff68NGzZo8eLFev755z3PBZwzZ4769eun3NxcTZ06VVOmTFH79u31xBNPaMKECZoxY4ZycnJO+5zf9PR0rV69Wp999plnWWlpqWbOnKl58+bp7bffVkREhPLy8kzbbwDnD4oagPNWz549JUnt2rVTZWWlfvjhB/Xq1UsBAQEKDw/3PCfy+++/15IlS5SWlqZJkybp2LFjkqQBAwZo//796tu3r9q1a3fabYSEhGjGjBmaOHGiysvLJUm7du1S165dFR4eLunEY5W2bt3a2LsL4DxEUQPgN2JjY7Vx40a53W4dP35cP/zwgySpS5cuGjlypHJzc/WnP/1JgwcPliS9/vrrSkhI0KZNm/Tll1/W+bmXXXaZBg0apFdffVWS1LFjR23btk3Hjx+XJK1fv16xsbGNvHcAzkdcowbAb/To0UO33HKLhgwZorZt26p169aSpFGjRumpp55Sfn6+HA6HHn74YX399df6xz/+oby8PO3atUtjx45VXl6eIiIiTvvZo0aN0ieffCJJio6O1tixY3XfffcpICBAMTExevzxx03bTwDnDx7KDgAAYFGc+gQAALAoihoAAIBFUdQAAAAsiqIGAABgURQ1AAAAi6KoAQAAWBRFDQAAwKIoagAAABb1/wHwq0JSL3TfiAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "in_out_degree_sort = sorted(in_out_degree)\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(in_out_degree_sort)\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of people each person is following + followers')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "in_out_degree_sort = sorted(in_out_degree)\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(in_out_degree_sort[0:1500000])\n",
    "plt.xlabel('Index No')\n",
    "plt.ylabel('No Of people each person is following + followers')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "90 percentile value is 24.0\n",
      "91 percentile value is 26.0\n",
      "92 percentile value is 28.0\n",
      "93 percentile value is 31.0\n",
      "94 percentile value is 33.0\n",
      "95 percentile value is 37.0\n",
      "96 percentile value is 41.0\n",
      "97 percentile value is 48.0\n",
      "98 percentile value is 58.0\n",
      "99 percentile value is 79.0\n",
      "100 percentile value is 1579.0\n"
     ]
    }
   ],
   "source": [
    "### 90-100 percentile\n",
    "for i in range(0,11):\n",
    "    print(90+i,'percentile value is',np.percentile(in_out_degree_sort,90+i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.1 percentile value is 83.0\n",
      "99.2 percentile value is 87.0\n",
      "99.3 percentile value is 93.0\n",
      "99.4 percentile value is 99.0\n",
      "99.5 percentile value is 108.0\n",
      "99.6 percentile value is 120.0\n",
      "99.7 percentile value is 138.0\n",
      "99.8 percentile value is 168.0\n",
      "99.9 percentile value is 221.0\n",
      "100.0 percentile value is 1579.0\n"
     ]
    }
   ],
   "source": [
    "### 99-100 percentile\n",
    "for i in range(10,110,10):\n",
    "    print(99+(i/100),'percentile value is',np.percentile(in_out_degree_sort,99+(i/100)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min of no of followers + following is 1\n",
      "334291  persons having minimum no of followers + following\n"
     ]
    }
   ],
   "source": [
    "print('Min of no of followers + following is',in_out_degree.min())\n",
    "print(np.sum(in_out_degree==in_out_degree.min()),' persons having minimum no of followers + following')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max of no of followers + following is 1579\n",
      "1  persons having maximum no of followers + following\n"
     ]
    }
   ],
   "source": [
    "print('Max of no of followers + following is',in_out_degree.max())\n",
    "print(np.sum(in_out_degree==in_out_degree.max()),' persons having maximum no of followers + following')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of persons having followers + following less than 10 are 1320326\n"
     ]
    }
   ],
   "source": [
    "print('No of persons having followers + following less than 10 are',np.sum(in_out_degree<10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of weakly connected components 45558\n",
      "weakly connected components wit 2 nodes 32195\n"
     ]
    }
   ],
   "source": [
    "print('No of weakly connected components',len(list(nx.weakly_connected_components(g))))\n",
    "count=0\n",
    "for i in list(nx.weakly_connected_components(g)):\n",
    "    if len(i)==2:\n",
    "        count+=1\n",
    "print('weakly connected components wit 2 nodes',count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Posing a problem as classification problem "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Generating some edges which are not present in graph for supervised learning  \n",
    "Generated Bad links from graph which are not in graph and whose shortest path is greater than 2. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 41.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "###generating bad edges from given graph\n",
    "import random\n",
    "if not os.path.isfile('data/after_eda/missing_edges_final.p'):\n",
    "    #getting all set of edges\n",
    "    r = csv.reader(open('data/after_eda/train_woheader.csv','r'))\n",
    "    edges = dict()\n",
    "    for edge in r:\n",
    "        edges[(edge[0], edge[1])] = 1\n",
    "        \n",
    "        \n",
    "    missing_edges = set([])\n",
    "    while (len(missing_edges)<9437519):\n",
    "        a=random.randint(1, 1862220)\n",
    "        b=random.randint(1, 1862220)\n",
    "        tmp = edges.get((a,b),-1)\n",
    "        if tmp == -1 and a!=b:\n",
    "            try:\n",
    "                if nx.shortest_path_length(g,source=a,target=b) > 2: \n",
    "\n",
    "                    missing_edges.add((a,b))\n",
    "                else:\n",
    "                    continue  \n",
    "            except:  \n",
    "                    missing_edges.add((a,b))              \n",
    "        else:\n",
    "            continue\n",
    "    pickle.dump(missing_edges,open('data/after_eda/missing_edges_final.p','wb'))\n",
    "else:\n",
    "    missing_edges = pickle.load(open('data/after_eda/missing_edges_final.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9437519"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(missing_edges)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.2 Training and Test data split:  \n",
    "Removed edges from Graph and used as test data and after removing used that graph for creating features for Train and test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of nodes in the graph with edges 9437519\n",
      "Number of nodes in the graph without edges 9437519\n",
      "============================================================\n",
      "Number of nodes in the train data graph with edges 7550015 = 7550015\n",
      "Number of nodes in the train data graph without edges 7550015 = 7550015\n",
      "============================================================\n",
      "Number of nodes in the test data graph with edges 1887504 = 1887504\n",
      "Number of nodes in the test data graph without edges 1887504 = 1887504\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "if (not os.path.isfile('data/after_eda/train_pos_after_eda.csv')) and (not os.path.isfile('data/after_eda/test_pos_after_eda.csv')):\n",
    "    #reading total data df\n",
    "    df_pos = pd.read_csv('data/train.csv')\n",
    "    df_neg = pd.DataFrame(list(missing_edges), columns=['source_node', 'destination_node'])\n",
    "    \n",
    "    print(\"Number of nodes in the graph with edges\", df_pos.shape[0])\n",
    "    print(\"Number of nodes in the graph without edges\", df_neg.shape[0])\n",
    "    \n",
    "    #Trian test split \n",
    "    #Spiltted data into 80-20 \n",
    "    #positive links and negative links seperatly because we need positive training data only for creating graph \n",
    "    #and for feature generation\n",
    "    X_train_pos, X_test_pos, y_train_pos, y_test_pos  = train_test_split(df_pos,np.ones(len(df_pos)),test_size=0.2, random_state=9)\n",
    "    X_train_neg, X_test_neg, y_train_neg, y_test_neg  = train_test_split(df_neg,np.zeros(len(df_neg)),test_size=0.2, random_state=9)\n",
    "    \n",
    "    print('='*60)\n",
    "    print(\"Number of nodes in the train data graph with edges\", X_train_pos.shape[0],\"=\",y_train_pos.shape[0])\n",
    "    print(\"Number of nodes in the train data graph without edges\", X_train_neg.shape[0],\"=\", y_train_neg.shape[0])\n",
    "    print('='*60)\n",
    "    print(\"Number of nodes in the test data graph with edges\", X_test_pos.shape[0],\"=\",y_test_pos.shape[0])\n",
    "    print(\"Number of nodes in the test data graph without edges\", X_test_neg.shape[0],\"=\",y_test_neg.shape[0])\n",
    "\n",
    "    #removing header and saving\n",
    "    X_train_pos.to_csv('data/after_eda/train_pos_after_eda.csv',header=False, index=False)\n",
    "    X_test_pos.to_csv('data/after_eda/test_pos_after_eda.csv',header=False, index=False)\n",
    "    X_train_neg.to_csv('data/after_eda/train_neg_after_eda.csv',header=False, index=False)\n",
    "    X_test_neg.to_csv('data/after_eda/test_neg_after_eda.csv',header=False, index=False)\n",
    "else:\n",
    "    #Graph from Traing data only \n",
    "    del missing_edges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: \n",
      "Type: DiGraph\n",
      "Number of nodes: 1780722\n",
      "Number of edges: 7550015\n",
      "Average in degree:   4.2399\n",
      "Average out degree:   4.2399\n",
      "Name: \n",
      "Type: DiGraph\n",
      "Number of nodes: 1144623\n",
      "Number of edges: 1887504\n",
      "Average in degree:   1.6490\n",
      "Average out degree:   1.6490\n",
      "no of people common in train and test --  1063125\n",
      "no of people present in train but not present in test --  717597\n",
      "no of people present in test but not present in train --  81498\n",
      " % of people not there in Train but exist in Test in total Test data are 7.1200735962845405 %\n"
     ]
    }
   ],
   "source": [
    "if (os.path.isfile('data/after_eda/train_pos_after_eda.csv')) and (os.path.isfile('data/after_eda/test_pos_after_eda.csv')):\n",
    "    train_graph=nx.read_edgelist('data/after_eda/train_pos_after_eda.csv',delimiter=',',create_using=nx.DiGraph(),nodetype=int)\n",
    "    test_graph=nx.read_edgelist('data/after_eda/test_pos_after_eda.csv',delimiter=',',create_using=nx.DiGraph(),nodetype=int)\n",
    "    print(nx.info(train_graph))\n",
    "    print(nx.info(test_graph))\n",
    "\n",
    "    # finding the unique nodes in the both train and test graphs\n",
    "    train_nodes_pos = set(train_graph.nodes())\n",
    "    test_nodes_pos = set(test_graph.nodes())\n",
    "\n",
    "    trY_teY = len(train_nodes_pos.intersection(test_nodes_pos))\n",
    "    trY_teN = len(train_nodes_pos - test_nodes_pos)\n",
    "    teY_trN = len(test_nodes_pos - train_nodes_pos)\n",
    "\n",
    "    print('no of people common in train and test -- ',trY_teY)\n",
    "    print('no of people present in train but not present in test -- ',trY_teN)\n",
    "\n",
    "    print('no of people present in test but not present in train -- ',teY_trN)\n",
    "    print(' % of people not there in Train but exist in Test in total Test data are {} %'.format(teY_trN/len(test_nodes_pos)*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> we have a cold start problem here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "============================================================\n",
      "Number of nodes in the train data graph with edges 7550015\n",
      "Number of nodes in the train data graph without edges 7550015\n",
      "============================================================\n",
      "Number of nodes in the test data graph with edges 1887504\n",
      "Number of nodes in the test data graph without edges 1887504\n"
     ]
    }
   ],
   "source": [
    "#final train and test data sets\n",
    "if (os.path.isfile('data/after_eda/train_pos_after_eda.csv')) and \\\n",
    "(os.path.isfile('data/after_eda/test_pos_after_eda.csv')) and \\\n",
    "(os.path.isfile('data/after_eda/train_neg_after_eda.csv')) and \\\n",
    "(os.path.isfile('data/after_eda/test_neg_after_eda.csv')):\n",
    "    \n",
    "    X_train_pos = pd.read_csv('data/after_eda/train_pos_after_eda.csv', names=['source_node', 'destination_node'])\n",
    "    X_test_pos = pd.read_csv('data/after_eda/test_pos_after_eda.csv', names=['source_node', 'destination_node'])\n",
    "    X_train_neg = pd.read_csv('data/after_eda/train_neg_after_eda.csv', names=['source_node', 'destination_node'])\n",
    "    X_test_neg = pd.read_csv('data/after_eda/test_neg_after_eda.csv', names=['source_node', 'destination_node'])\n",
    "\n",
    "    print('='*60)\n",
    "    print(\"Number of nodes in the train data graph with edges\", X_train_pos.shape[0])\n",
    "    print(\"Number of nodes in the train data graph without edges\", X_train_neg.shape[0])\n",
    "    print('='*60)\n",
    "    print(\"Number of nodes in the test data graph with edges\", X_test_pos.shape[0])\n",
    "    print(\"Number of nodes in the test data graph without edges\", X_test_neg.shape[0])\n",
    "\n",
    "    X_train = X_train_pos.append(X_train_neg,ignore_index=True)\n",
    "    y_train = np.concatenate((y_train_pos,y_train_neg))\n",
    "    X_test = X_test_pos.append(X_test_neg,ignore_index=True)\n",
    "    y_test = np.concatenate((y_test_pos,y_test_neg)) \n",
    "    \n",
    "    X_train.to_csv('data/after_eda/train_after_eda.csv',header=False,index=False)\n",
    "    X_test.to_csv('data/after_eda/test_after_eda.csv',header=False,index=False)\n",
    "    pd.DataFrame(y_train.astype(int)).to_csv('data/train_y.csv',header=False,index=False)\n",
    "    pd.DataFrame(y_test.astype(int)).to_csv('data/test_y.csv',header=False,index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data points in train data (15100030, 2)\n",
      "Data points in test data (3775008, 2)\n",
      "Shape of traget variable in train (15100030,)\n",
      "Shape of traget variable in test (3775008,)\n"
     ]
    }
   ],
   "source": [
    "print(\"Data points in train data\",X_train.shape)\n",
    "print(\"Data points in test data\",X_test.shape)\n",
    "print(\"Shape of traget variable in train\",y_train.shape)\n",
    "print(\"Shape of traget variable in test\", y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# FB_featurization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Importing Libraries \n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import csv\n",
    "import pandas as pd#pandas to create small dataframes \n",
    "import datetime #Convert to unix time\n",
    "import time #Convert to unix time\n",
    "# if numpy is not installed already : pip3 install numpy\n",
    "import numpy as np#Do aritmetic operations on arrays\n",
    "# matplotlib: used to plot graphs\n",
    "import matplotlib\n",
    "import matplotlib.pylab as plt\n",
    "import seaborn as sns#Plots\n",
    "from matplotlib import rcParams#Size of plots  \n",
    "from sklearn.cluster import MiniBatchKMeans, KMeans#Clustering\n",
    "import math\n",
    "import pickle\n",
    "import os\n",
    "# to install xgboost: pip3 install xgboost\n",
    "import xgboost as xgb\n",
    "\n",
    "import warnings\n",
    "import networkx as nx\n",
    "import pdb\n",
    "import pickle\n",
    "\n",
    "\n",
    "from pandas import HDFStore,DataFrame\n",
    "from pandas import read_hdf\n",
    "from scipy.sparse.linalg import svds, eigs\n",
    "import gc\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Reading Data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: \n",
      "Type: DiGraph\n",
      "Number of nodes: 1780722\n",
      "Number of edges: 7550015\n",
      "Average in degree:   4.2399\n",
      "Average out degree:   4.2399\n"
     ]
    }
   ],
   "source": [
    "if os.path.isfile('data/after_eda/train_pos_after_eda.csv'):\n",
    "    train_graph=nx.read_edgelist('data/after_eda/train_pos_after_eda.csv',delimiter=',',create_using=nx.DiGraph(),nodetype=int)\n",
    "    print(nx.info(train_graph))\n",
    "else:\n",
    "    print(\"please create the file\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Similarity measures"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Jaccard Distance: \n",
    "http://www.statisticshowto.com/jaccard-index/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{equation}\n",
    "j = \\frac{|X\\cap Y|}{|X \\cup Y|} \n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for followees\n",
    "def jaccard_for_followees(a,b):\n",
    "    try:\n",
    "        if len(set(train_graph.successors(a))) == 0  | len(set(train_graph.successors(b))) == 0:\n",
    "            return 0\n",
    "        sim = (len(set(train_graph.successors(a)).intersection(set(train_graph.successors(b)))))/\\\n",
    "                                    (len(set(train_graph.successors(a)).union(set(train_graph.successors(b)))))\n",
    "    except:\n",
    "        return 0\n",
    "    return sim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    }
   ],
   "source": [
    "#one test case\n",
    "print(jaccard_for_followees(273084,1505602))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    }
   ],
   "source": [
    "#node 1635354 not in graph \n",
    "print(jaccard_for_followees(273084,1505602))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for followers\n",
    "def jaccard_for_followers(a,b):\n",
    "    try:\n",
    "        if len(set(train_graph.predecessors(a))) == 0  | len(set(g.predecessors(b))) == 0:\n",
    "            return 0\n",
    "        sim = (len(set(train_graph.predecessors(a)).intersection(set(train_graph.predecessors(b)))))/\\\n",
    "                                 (len(set(train_graph.predecessors(a)).union(set(train_graph.predecessors(b)))))\n",
    "        return sim\n",
    "    except:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "print(jaccard_for_followers(273084,470294))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "#node 1635354 not in graph \n",
    "print(jaccard_for_followees(669354,1635354))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.2 Cosine distance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{equation}\n",
    "CosineDistance = \\frac{|X\\cap Y|}{|X|\\cdot|Y|} \n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for followees\n",
    "def cosine_for_followees(a,b):\n",
    "    try:\n",
    "        if len(set(train_graph.successors(a))) == 0  | len(set(train_graph.successors(b))) == 0:\n",
    "            return 0\n",
    "        sim = (len(set(train_graph.successors(a)).intersection(set(train_graph.successors(b)))))/\\\n",
    "                                    (math.sqrt(len(set(train_graph.successors(a)))*len((set(train_graph.successors(b))))))\n",
    "        return sim\n",
    "    except:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    }
   ],
   "source": [
    "print(cosine_for_followees(273084,1505602))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "print(cosine_for_followees(273084,1635354))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cosine_for_followers(a,b):\n",
    "    try:\n",
    "        if len(set(train_graph.predecessors(a))) == 0  | len(set(train_graph.predecessors(b))) == 0:\n",
    "            return 0\n",
    "        sim = (len(set(train_graph.predecessors(a)).intersection(set(train_graph.predecessors(b)))))/\\\n",
    "                                     (math.sqrt(len(set(train_graph.predecessors(a))))*(len(set(train_graph.predecessors(b)))))\n",
    "        return sim\n",
    "    except:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.02886751345948129\n"
     ]
    }
   ],
   "source": [
    "print(cosine_for_followers(2,470294))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "print(cosine_for_followers(669354,1635354))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Ranking Measures "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://networkx.github.io/documentation/networkx-1.10/reference/generated/networkx.algorithms.link_analysis.pagerank_alg.pagerank.html\n",
    "\n",
    "PageRank computes a ranking of the nodes in the graph G based on the structure of the incoming links.\n",
    "\n",
    "<img src='PageRanks-Example.jpg'/>\n",
    "\n",
    "Mathematical PageRanks for a simple network, expressed as percentages. (Google uses a logarithmic scale.) Page C has a higher PageRank than Page E, even though there are fewer links to C; the one link to C comes from an important page and hence is of high value. If web surfers who start on a random page have an 85% likelihood of choosing a random link from the page they are currently visiting, and a 15% likelihood of jumping to a page chosen at random from the entire web, they will reach Page E 8.1% of the time. <b>(The 15% likelihood of jumping to an arbitrary page corresponds to a damping factor of 85%.) Without damping, all web surfers would eventually end up on Pages A, B, or C, and all other pages would have PageRank zero. In the presence of damping, Page A effectively links to all pages in the web, even though it has no outgoing links of its own.</b>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Page Ranking\n",
    "\n",
    "https://en.wikipedia.org/wiki/PageRank "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/page_rank.p'):\n",
    "    pr = nx.pagerank(train_graph, alpha=0.85)\n",
    "    pickle.dump(pr,open('data/fea_sample/page_rank.p','wb'))\n",
    "else:\n",
    "    pr = pickle.load(open('data/fea_sample/page_rank.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min 1.6556497245737814e-07\n",
      "max 2.7098251341935827e-05\n",
      "mean 5.615699699365892e-07\n"
     ]
    }
   ],
   "source": [
    "print('min',pr[min(pr, key=pr.get)])\n",
    "print('max',pr[max(pr, key=pr.get)])\n",
    "print('mean',float(sum(pr.values())) / len(pr))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.615699699365892e-07\n"
     ]
    }
   ],
   "source": [
    "#for imputing to nodes which are not there in Train data\n",
    "mean_pr = float(sum(pr.values())) / len(pr)\n",
    "print(mean_pr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Other Graph Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Shortest path: "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting Shortest path between twoo nodes, if nodes have direct path i.e directly connected then we are removing that edge and calculating path."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#if has direct edge then deleting that edge and calculating shortest path\n",
    "def compute_shortest_path_length(a,b):\n",
    "    p=-1\n",
    "    try:\n",
    "        if train_graph.has_edge(a,b):\n",
    "            train_graph.remove_edge(a,b)\n",
    "            p= nx.shortest_path_length(train_graph,source=a,target=b)\n",
    "            train_graph.add_edge(a,b)\n",
    "        else:\n",
    "            p= nx.shortest_path_length(train_graph,source=a,target=b)\n",
    "        return p\n",
    "    except:\n",
    "        return -1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#testing\n",
    "compute_shortest_path_length(77697, 826021)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#testing\n",
    "compute_shortest_path_length(669354,1635354)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Checking for same community "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#getting weekly connected edges from graph \n",
    "wcc=list(nx.weakly_connected_components(train_graph))\n",
    "def belongs_to_same_wcc(a,b):\n",
    "    index = []\n",
    "    if train_graph.has_edge(b,a):\n",
    "        return 1\n",
    "    if train_graph.has_edge(a,b):\n",
    "            for i in wcc:\n",
    "                if a in i:\n",
    "                    index= i\n",
    "                    break\n",
    "            if (b in index):\n",
    "                train_graph.remove_edge(a,b)\n",
    "                if compute_shortest_path_length(a,b)==-1:\n",
    "                    train_graph.add_edge(a,b)\n",
    "                    return 0\n",
    "                else:\n",
    "                    train_graph.add_edge(a,b)\n",
    "                    return 1\n",
    "            else:\n",
    "                return 0\n",
    "    else:\n",
    "            for i in wcc:\n",
    "                if a in i:\n",
    "                    index= i\n",
    "                    break\n",
    "            if(b in index):\n",
    "                return 1\n",
    "            else:\n",
    "                return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "belongs_to_same_wcc(861, 1659750)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "belongs_to_same_wcc(669354,1635354)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.3 Adamic/Adar Index: "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adamic/Adar measures is defined as inverted sum of degrees of common neighbours for given two vertices.\n",
    "$$A(x,y)=\\sum_{u \\in N(x) \\cap N(y)}\\frac{1}{log(|N(u)|)}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#adar index\n",
    "def calc_adar_in(a,b):\n",
    "    sum=0\n",
    "    try:\n",
    "        n=list(set(train_graph.successors(a)).intersection(set(train_graph.successors(b))))\n",
    "        if len(n)!=0:\n",
    "            for i in n:\n",
    "                sum=sum+(1/np.log10(len(list(train_graph.predecessors(i)))))\n",
    "            return sum\n",
    "        else:\n",
    "            return 0\n",
    "    except:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "calc_adar_in(1,189226)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "calc_adar_in(669354,1635354)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.4 Is person was following back: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def follows_back(a,b):\n",
    "    if train_graph.has_edge(b,a):\n",
    "        return 1\n",
    "    else:\n",
    "        return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "follows_back(1,189226)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "follows_back(669354,1635354)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.5 Katz Centrality:\n",
    "https://en.wikipedia.org/wiki/Katz_centrality\n",
    "\n",
    "https://www.geeksforgeeks.org/katz-centrality-centrality-measure/\n",
    " Katz centrality computes the centrality for a node \n",
    "    based on the centrality of its neighbors. It is a \n",
    "    generalization of the eigenvector centrality. The\n",
    "    Katz centrality for node `i` is\n",
    " \n",
    "$$x_i = \\alpha \\sum_{j} A_{ij} x_j + \\beta,$$\n",
    "where `A` is the adjacency matrix of the graph G \n",
    "with eigenvalues $$\\lambda$$.\n",
    "\n",
    "The parameter $$\\beta$$ controls the initial centrality and \n",
    "\n",
    "$$\\alpha < \\frac{1}{\\lambda_{max}}.$$ "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/katz.p'):\n",
    "    katz = nx.katz.katz_centrality(train_graph,alpha=0.005,beta=1)\n",
    "    pickle.dump(katz,open('data/fea_sample/katz.p','wb'))\n",
    "else:\n",
    "    katz = pickle.load(open('data/fea_sample/katz.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min 0.0007313532484065916\n",
      "max 0.003394554981699122\n",
      "mean 0.0007483800935504637\n"
     ]
    }
   ],
   "source": [
    "print('min',katz[min(katz, key=katz.get)])\n",
    "print('max',katz[max(katz, key=katz.get)])\n",
    "print('mean',float(sum(katz.values())) / len(katz))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0007483800935504637\n"
     ]
    }
   ],
   "source": [
    "mean_katz = float(sum(katz.values())) / len(katz)\n",
    "print(mean_katz)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.6 Hits Score\n",
    "The HITS algorithm computes two numbers for a node. Authorities estimates the node value based on the incoming links. Hubs estimates the node value based on outgoing links.\n",
    "\n",
    "https://en.wikipedia.org/wiki/HITS_algorithm "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/hits.p'):\n",
    "    hits = nx.hits(train_graph, max_iter=100, tol=1e-08, nstart=None, normalized=True)\n",
    "    pickle.dump(hits,open('data/fea_sample/hits.p','wb'))\n",
    "else:\n",
    "    hits = pickle.load(open('data/fea_sample/hits.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min 0.0\n",
      "max 0.004868653378780953\n",
      "mean 5.615699699353278e-07\n"
     ]
    }
   ],
   "source": [
    "print('min',hits[0][min(hits[0], key=hits[0].get)])\n",
    "print('max',hits[0][max(hits[0], key=hits[0].get)])\n",
    "print('mean',float(sum(hits[0].values())) / len(hits[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Featurization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.1 Reading a sample of Data from both train and test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "if os.path.isfile('data/after_eda/train_after_eda.csv'):\n",
    "    filename = \"data/after_eda/train_after_eda.csv\"\n",
    "    # you uncomment this line, if you dont know the lentgh of the file name\n",
    "    # here we have hardcoded the number of lines as 15100030\n",
    "    # n_train = sum(1 for line in open(filename)) #number of records in file (excludes header)\n",
    "    n_train =  15100028\n",
    "    s = 100000 #desired sample size\n",
    "    skip_train = sorted(random.sample(range(1,n_train+1),n_train-s))\n",
    "    #https://stackoverflow.com/a/22259008/4084039"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "if os.path.isfile('data/after_eda/test_after_eda.csv'):\n",
    "    filename = \"data/after_eda/test_after_eda.csv\"\n",
    "    # you uncomment this line, if you dont know the lentgh of the file name\n",
    "    # here we have hardcoded the number of lines as 3775008\n",
    "    # n_test = sum(1 for line in open(filename)) #number of records in file (excludes header)\n",
    "    n_test = 3775006\n",
    "    s = 50000 #desired sample size\n",
    "    skip_test = sorted(random.sample(range(1,n_test+1),n_test-s))\n",
    "    #https://stackoverflow.com/a/22259008/4084039"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows in the train data file: 15100028\n",
      "Number of rows we are going to elimiate in train data are 15000028\n",
      "Number of rows in the test data file: 3775006\n",
      "Number of rows we are going to elimiate in test data are 3725006\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of rows in the train data file:\", n_train)\n",
    "print(\"Number of rows we are going to elimiate in train data are\",len(skip_train))\n",
    "print(\"Number of rows in the test data file:\", n_test)\n",
    "print(\"Number of rows we are going to elimiate in test data are\",len(skip_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Our train matrix size  (100002, 3)\n"
     ]
    },
    {
     "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link\n",
       "0       273084           1505602               1\n",
       "1        38688            353776               1"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train = pd.read_csv('data/after_eda/train_after_eda.csv', skiprows=skip_train, names=['source_node', 'destination_node'])\n",
    "df_final_train['indicator_link'] = pd.read_csv('data/train_y.csv', skiprows=skip_train, names=['indicator_link'])\n",
    "print(\"Our train matrix size \",df_final_train.shape)\n",
    "df_final_train.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Our test matrix size  (50002, 3)\n"
     ]
    },
    {
     "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>848424</td>\n",
       "      <td>784690</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>740824</td>\n",
       "      <td>956372</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link\n",
       "0       848424            784690               1\n",
       "1       740824            956372               1"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_test = pd.read_csv('data/after_eda/test_after_eda.csv', skiprows=skip_test, names=['source_node', 'destination_node'])\n",
    "df_final_test['indicator_link'] = pd.read_csv('data/test_y.csv', skiprows=skip_test, names=['indicator_link'])\n",
    "print(\"Our test matrix size \",df_final_test.shape)\n",
    "df_final_test.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.2 Adding a set of features\n",
    "\n",
    "__we will create these each of these features for both train and test data points__\n",
    "<ol>\n",
    "<li>jaccard_followers</li>\n",
    "<li>jaccard_followees</li>\n",
    "<li>cosine_followers</li>\n",
    "<li>cosine_followees</li>\n",
    "<li>num_followers_s</li>\n",
    "<li>num_followees_s</li>\n",
    "<li>num_followers_d</li>\n",
    "<li>num_followees_d</li>\n",
    "<li>inter_followers</li>\n",
    "<li>inter_followees</li>\n",
    "</ol>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage1.h5'):\n",
    "    #mapping jaccrd followers to train and test data\n",
    "    df_final_train['jaccard_followers'] = df_final_train.apply(lambda row:\n",
    "                                            jaccard_for_followers(row['source_node'],row['destination_node']),axis=1)\n",
    "    df_final_test['jaccard_followers'] = df_final_test.apply(lambda row:\n",
    "                                            jaccard_for_followers(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    #mapping jaccrd followees to train and test data\n",
    "    df_final_train['jaccard_followees'] = df_final_train.apply(lambda row:\n",
    "                                            jaccard_for_followees(row['source_node'],row['destination_node']),axis=1)\n",
    "    df_final_test['jaccard_followees'] = df_final_test.apply(lambda row:\n",
    "                                            jaccard_for_followees(row['source_node'],row['destination_node']),axis=1)\n",
    "    \n",
    "\n",
    "        #mapping jaccrd followers to train and test data\n",
    "    df_final_train['cosine_followers'] = df_final_train.apply(lambda row:\n",
    "                                            cosine_for_followers(row['source_node'],row['destination_node']),axis=1)\n",
    "    df_final_test['cosine_followers'] = df_final_test.apply(lambda row:\n",
    "                                            cosine_for_followers(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    #mapping jaccrd followees to train and test data\n",
    "    df_final_train['cosine_followees'] = df_final_train.apply(lambda row:\n",
    "                                            cosine_for_followees(row['source_node'],row['destination_node']),axis=1)\n",
    "    df_final_test['cosine_followees'] = df_final_test.apply(lambda row:\n",
    "                                            cosine_for_followees(row['source_node'],row['destination_node']),axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_features_stage1(df_final):\n",
    "    #calculating no of followers followees for source and destination\n",
    "    #calculating intersection of followers and followees for source and destination\n",
    "    num_followers_s=[]\n",
    "    num_followees_s=[]\n",
    "    num_followers_d=[]\n",
    "    num_followees_d=[]\n",
    "    inter_followers=[]\n",
    "    inter_followees=[]\n",
    "    for i,row in df_final.iterrows():\n",
    "        try:\n",
    "            s1=set(train_graph.predecessors(row['source_node']))\n",
    "            s2=set(train_graph.successors(row['source_node']))\n",
    "        except:\n",
    "            s1 = set()\n",
    "            s2 = set()\n",
    "        try:\n",
    "            d1=set(train_graph.predecessors(row['destination_node']))\n",
    "            d2=set(train_graph.successors(row['destination_node']))\n",
    "        except:\n",
    "            d1 = set()\n",
    "            d2 = set()\n",
    "        num_followers_s.append(len(s1))\n",
    "        num_followees_s.append(len(s2))\n",
    "\n",
    "        num_followers_d.append(len(d1))\n",
    "        num_followees_d.append(len(d2))\n",
    "\n",
    "        inter_followers.append(len(s1.intersection(d1)))\n",
    "        inter_followees.append(len(s2.intersection(d2)))\n",
    "    \n",
    "    return num_followers_s, num_followers_d, num_followees_s, num_followees_d, inter_followers, inter_followees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage1.h5'):\n",
    "    df_final_train['num_followers_s'], df_final_train['num_followers_d'], \\\n",
    "    df_final_train['num_followees_s'], df_final_train['num_followees_d'], \\\n",
    "    df_final_train['inter_followers'], df_final_train['inter_followees']= compute_features_stage1(df_final_train)\n",
    "    \n",
    "    df_final_test['num_followers_s'], df_final_test['num_followers_d'], \\\n",
    "    df_final_test['num_followees_s'], df_final_test['num_followees_d'], \\\n",
    "    df_final_test['inter_followers'], df_final_test['inter_followees']= compute_features_stage1(df_final_test)\n",
    "    \n",
    "    hdf = HDFStore('data/fea_sample/storage_sample_stage1.h5')\n",
    "    hdf.put('train_df',df_final_train, format='table', data_columns=True)\n",
    "    hdf.put('test_df',df_final_test, format='table', data_columns=True)\n",
    "    hdf.close()\n",
    "else:\n",
    "    df_final_train = read_hdf('data/fea_sample/storage_sample_stage1.h5', 'train_df',mode='r')\n",
    "    df_final_test = read_hdf('data/fea_sample/storage_sample_stage1.h5', 'test_df',mode='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100002, 13)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(50002, 13)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df_final_train.shape)\n",
    "df_final_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['source_node', 'destination_node', 'indicator_link',\n",
       "       'jaccard_followers', 'jaccard_followees', 'cosine_followers',\n",
       "       'cosine_followees', 'num_followers_s', 'num_followers_d',\n",
       "       'num_followees_s', 'num_followees_d', 'inter_followers',\n",
       "       'inter_followees'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.3 Adding new set of features\n",
    "\n",
    "__we will create these each of these features for both train and test data points__\n",
    "<ol>\n",
    "<li>adar index</li>\n",
    "<li>is following back</li>\n",
    "<li>belongs to same weakly connect components</li>\n",
    "<li>shortest path between source and destination</li>\n",
    "</ol>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage2.h5'):\n",
    "    #mapping adar index on train\n",
    "    df_final_train['adar_index'] = df_final_train.apply(lambda row: calc_adar_in(row['source_node'],row['destination_node']),axis=1)\n",
    "    #mapping adar index on test\n",
    "    df_final_test['adar_index'] = df_final_test.apply(lambda row: calc_adar_in(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    #--------------------------------------------------------------------------------------------------------\n",
    "    #mapping followback or not on train\n",
    "    df_final_train['follows_back'] = df_final_train.apply(lambda row: follows_back(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    #mapping followback or not on test\n",
    "    df_final_test['follows_back'] = df_final_test.apply(lambda row: follows_back(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    #--------------------------------------------------------------------------------------------------------\n",
    "    #mapping same component of wcc or not on train\n",
    "    df_final_train['same_comp'] = df_final_train.apply(lambda row: belongs_to_same_wcc(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    ##mapping same component of wcc or not on train\n",
    "    df_final_test['same_comp'] = df_final_test.apply(lambda row: belongs_to_same_wcc(row['source_node'],row['destination_node']),axis=1)\n",
    "    \n",
    "    #--------------------------------------------------------------------------------------------------------\n",
    "    #mapping shortest path on train \n",
    "    df_final_train['shortest_path'] = df_final_train.apply(lambda row: compute_shortest_path_length(row['source_node'],row['destination_node']),axis=1)\n",
    "    #mapping shortest path on test\n",
    "    df_final_test['shortest_path'] = df_final_test.apply(lambda row: compute_shortest_path_length(row['source_node'],row['destination_node']),axis=1)\n",
    "\n",
    "    hdf = HDFStore('data/fea_sample/storage_sample_stage2.h5')\n",
    "    hdf.put('train_df',df_final_train, format='table', data_columns=True)\n",
    "    hdf.put('test_df',df_final_test, format='table', data_columns=True)\n",
    "    hdf.close()\n",
    "else:\n",
    "    df_final_train = read_hdf('data/fea_sample/storage_sample_stage2.h5', 'train_df',mode='r')\n",
    "    df_final_test = read_hdf('data/fea_sample/storage_sample_stage2.h5', 'test_df',mode='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100002, 17)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(50002, 17)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df_final_train.shape)\n",
    "df_final_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.4 Adding new set of features\n",
    "\n",
    "__we will create these each of these features for both train and test data points__\n",
    "<ol>\n",
    "<li>Weight Features\n",
    "    <ul>\n",
    "        <li>weight of incoming edges</li>\n",
    "        <li>weight of outgoing edges</li>\n",
    "        <li>weight of incoming edges + weight of outgoing edges</li>\n",
    "        <li>weight of incoming edges * weight of outgoing edges</li>\n",
    "        <li>2*weight of incoming edges + weight of outgoing edges</li>\n",
    "        <li>weight of incoming edges + 2*weight of outgoing edges</li>\n",
    "    </ul>\n",
    "</li>\n",
    "<li>Page Ranking of source</li>\n",
    "<li>Page Ranking of dest</li>\n",
    "<li>katz of source</li>\n",
    "<li>katz of dest</li>\n",
    "<li>hubs of source</li>\n",
    "<li>hubs of dest</li>\n",
    "<li>authorities_s of source</li>\n",
    "<li>authorities_s of dest</li>\n",
    "</ol> "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Weight Features\n",
    "In order to determine the similarity of nodes, an edge weight value was calculated between nodes. Edge weight decreases as the neighbor count goes up. Intuitively, consider one million people following a celebrity on a social network then chances are most of them never met each other or the celebrity. On the other hand, if a user has 30 contacts in his/her social network, the chances are higher that many of them know each other.  credit - Graph-based Features for Supervised Link Prediction William Cukierski, Benjamin Hamner, Bo Yang"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{equation}\n",
    "W = \\frac{1}{\\sqrt{1+|X|}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is directed graph so calculated Weighted in and Weighted out differently"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████| 1780722/1780722 [00:31<00:00, 56566.42it/s]\n"
     ]
    }
   ],
   "source": [
    "#weight for source and destination of each link\n",
    "Weight_in = {}\n",
    "Weight_out = {}\n",
    "for i in  tqdm(train_graph.nodes()):\n",
    "    s1=set(train_graph.predecessors(i))\n",
    "    w_in = 1.0/(np.sqrt(1+len(s1)))\n",
    "    Weight_in[i]=w_in\n",
    "    \n",
    "    s2=set(train_graph.successors(i))\n",
    "    w_out = 1.0/(np.sqrt(1+len(s2)))\n",
    "    Weight_out[i]=w_out\n",
    "    \n",
    "#for imputing with mean\n",
    "mean_weight_in = np.mean(list(Weight_in.values()))\n",
    "mean_weight_out = np.mean(list(Weight_out.values()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage3.h5'):\n",
    "    #mapping to pandas train\n",
    "    df_final_train['weight_in'] = df_final_train.destination_node.apply(lambda x: Weight_in.get(x,mean_weight_in))\n",
    "    df_final_train['weight_out'] = df_final_train.source_node.apply(lambda x: Weight_out.get(x,mean_weight_out))\n",
    "\n",
    "    #mapping to pandas test\n",
    "    df_final_test['weight_in'] = df_final_test.destination_node.apply(lambda x: Weight_in.get(x,mean_weight_in))\n",
    "    df_final_test['weight_out'] = df_final_test.source_node.apply(lambda x: Weight_out.get(x,mean_weight_out))\n",
    "\n",
    "\n",
    "    #some features engineerings on the in and out weights\n",
    "    df_final_train['weight_f1'] = df_final_train.weight_in + df_final_train.weight_out\n",
    "    df_final_train['weight_f2'] = df_final_train.weight_in * df_final_train.weight_out\n",
    "    df_final_train['weight_f3'] = (2*df_final_train.weight_in + 1*df_final_train.weight_out)\n",
    "    df_final_train['weight_f4'] = (1*df_final_train.weight_in + 2*df_final_train.weight_out)\n",
    "\n",
    "    #some features engineerings on the in and out weights\n",
    "    df_final_test['weight_f1'] = df_final_test.weight_in + df_final_test.weight_out\n",
    "    df_final_test['weight_f2'] = df_final_test.weight_in * df_final_test.weight_out\n",
    "    df_final_test['weight_f3'] = (2*df_final_test.weight_in + 1*df_final_test.weight_out)\n",
    "    df_final_test['weight_f4'] = (1*df_final_test.weight_in + 2*df_final_test.weight_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage3.h5'):\n",
    "    \n",
    "    #page rank for source and destination in Train and Test\n",
    "    #if anything not there in train graph then adding mean page rank \n",
    "    df_final_train['page_rank_s'] = df_final_train.source_node.apply(lambda x:pr.get(x,mean_pr))\n",
    "    df_final_train['page_rank_d'] = df_final_train.destination_node.apply(lambda x:pr.get(x,mean_pr))\n",
    "\n",
    "    df_final_test['page_rank_s'] = df_final_test.source_node.apply(lambda x:pr.get(x,mean_pr))\n",
    "    df_final_test['page_rank_d'] = df_final_test.destination_node.apply(lambda x:pr.get(x,mean_pr))\n",
    "    #================================================================================\n",
    "\n",
    "    #Katz centrality score for source and destination in Train and test\n",
    "    #if anything not there in train graph then adding mean katz score\n",
    "    df_final_train['katz_s'] = df_final_train.source_node.apply(lambda x: katz.get(x,mean_katz))\n",
    "    df_final_train['katz_d'] = df_final_train.destination_node.apply(lambda x: katz.get(x,mean_katz))\n",
    "\n",
    "    df_final_test['katz_s'] = df_final_test.source_node.apply(lambda x: katz.get(x,mean_katz))\n",
    "    df_final_test['katz_d'] = df_final_test.destination_node.apply(lambda x: katz.get(x,mean_katz))\n",
    "    #================================================================================\n",
    "\n",
    "    #Hits algorithm score for source and destination in Train and test\n",
    "    #if anything not there in train graph then adding 0\n",
    "    df_final_train['hubs_s'] = df_final_train.source_node.apply(lambda x: hits[0].get(x,0))\n",
    "    df_final_train['hubs_d'] = df_final_train.destination_node.apply(lambda x: hits[0].get(x,0))\n",
    "\n",
    "    df_final_test['hubs_s'] = df_final_test.source_node.apply(lambda x: hits[0].get(x,0))\n",
    "    df_final_test['hubs_d'] = df_final_test.destination_node.apply(lambda x: hits[0].get(x,0))\n",
    "    #================================================================================\n",
    "\n",
    "    #Hits algorithm score for source and destination in Train and Test\n",
    "    #if anything not there in train graph then adding 0\n",
    "    df_final_train['authorities_s'] = df_final_train.source_node.apply(lambda x: hits[1].get(x,0))\n",
    "    df_final_train['authorities_d'] = df_final_train.destination_node.apply(lambda x: hits[1].get(x,0))\n",
    "\n",
    "    df_final_test['authorities_s'] = df_final_test.source_node.apply(lambda x: hits[1].get(x,0))\n",
    "    df_final_test['authorities_d'] = df_final_test.destination_node.apply(lambda x: hits[1].get(x,0))\n",
    "    #================================================================================\n",
    "\n",
    "    hdf = HDFStore('data/fea_sample/storage_sample_stage3.h5')\n",
    "    hdf.put('train_df',df_final_train, format='table', data_columns=True)\n",
    "    hdf.put('test_df',df_final_test, format='table', data_columns=True)\n",
    "    hdf.close()\n",
    "else:\n",
    "    df_final_train = read_hdf('data/fea_sample/storage_sample_stage3.h5', 'train_df',mode='r')\n",
    "    df_final_test = read_hdf('data/fea_sample/storage_sample_stage3.h5', 'test_df',mode='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>weight_f3</th>\n",
       "      <th>weight_f4</th>\n",
       "      <th>page_rank_s</th>\n",
       "      <th>page_rank_d</th>\n",
       "      <th>katz_s</th>\n",
       "      <th>katz_d</th>\n",
       "      <th>hubs_s</th>\n",
       "      <th>hubs_d</th>\n",
       "      <th>authorities_s</th>\n",
       "      <th>authorities_d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>1.005929</td>\n",
       "      <td>0.877964</td>\n",
       "      <td>2.045290e-06</td>\n",
       "      <td>3.459963e-07</td>\n",
       "      <td>0.000773</td>\n",
       "      <td>0.000756</td>\n",
       "      <td>1.943132e-13</td>\n",
       "      <td>1.941103e-13</td>\n",
       "      <td>9.226339e-16</td>\n",
       "      <td>2.231877e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.050252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>1.089262</td>\n",
       "      <td>1.044631</td>\n",
       "      <td>7.593942e-07</td>\n",
       "      <td>4.069391e-07</td>\n",
       "      <td>0.000774</td>\n",
       "      <td>0.000755</td>\n",
       "      <td>5.943448e-16</td>\n",
       "      <td>1.871649e-14</td>\n",
       "      <td>2.075432e-15</td>\n",
       "      <td>8.766780e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>753961</td>\n",
       "      <td>1843752</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.042322e-07</td>\n",
       "      <td>6.428994e-07</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>4.764691e-18</td>\n",
       "      <td>3.350644e-20</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       273084           1505602               1                  0   \n",
       "1        38688            353776               1                  0   \n",
       "2       753961           1843752               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0                0.0          0.000000               0.0               11   \n",
       "1                0.0          0.050252               0.0               11   \n",
       "2                0.0          0.000000               0.0                1   \n",
       "\n",
       "   num_followers_d  num_followees_s      ...        weight_f3  weight_f4  \\\n",
       "0                6               15      ...         1.005929   0.877964   \n",
       "1                6                8      ...         1.089262   1.044631   \n",
       "2                1                1      ...         3.000000   3.000000   \n",
       "\n",
       "    page_rank_s   page_rank_d    katz_s    katz_d        hubs_s        hubs_d  \\\n",
       "0  2.045290e-06  3.459963e-07  0.000773  0.000756  1.943132e-13  1.941103e-13   \n",
       "1  7.593942e-07  4.069391e-07  0.000774  0.000755  5.943448e-16  1.871649e-14   \n",
       "2  4.042322e-07  6.428994e-07  0.000735  0.000735  0.000000e+00  4.764691e-18   \n",
       "\n",
       "   authorities_s  authorities_d  \n",
       "0   9.226339e-16   2.231877e-15  \n",
       "1   2.075432e-15   8.766780e-16  \n",
       "2   3.350644e-20   0.000000e+00  \n",
       "\n",
       "[3 rows x 31 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['source_node', 'destination_node', 'indicator_link',\n",
       "       'jaccard_followers', 'jaccard_followees', 'cosine_followers',\n",
       "       'cosine_followees', 'num_followers_s', 'num_followers_d',\n",
       "       'num_followees_s', 'num_followees_d', 'inter_followers',\n",
       "       'inter_followees', 'adar_index', 'follows_back', 'same_comp',\n",
       "       'shortest_path', 'weight_in', 'weight_out', 'weight_f1', 'weight_f2',\n",
       "       'weight_f3', 'weight_f4', 'page_rank_s', 'page_rank_d', 'katz_s',\n",
       "       'katz_d', 'hubs_s', 'hubs_d', 'authorities_s', 'authorities_d'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adding new feature called Preferential Attachement "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One well-known concept in social networks is that users with many friends tend to create more connections in the future. This is due to the fact that in some social networks, like in finance, the rich get richer. We estimate how ”rich” our two vertices are by calculating the multiplication between the number of friends (|Γ(x)|) or followers each vertex has."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preferential Attachement for followers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100002, 31)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>weight_f4</th>\n",
       "      <th>page_rank_s</th>\n",
       "      <th>page_rank_d</th>\n",
       "      <th>katz_s</th>\n",
       "      <th>katz_d</th>\n",
       "      <th>hubs_s</th>\n",
       "      <th>hubs_d</th>\n",
       "      <th>authorities_s</th>\n",
       "      <th>authorities_d</th>\n",
       "      <th>prefer_Attach_followers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>0.877964</td>\n",
       "      <td>2.045290e-06</td>\n",
       "      <td>3.459963e-07</td>\n",
       "      <td>0.000773</td>\n",
       "      <td>0.000756</td>\n",
       "      <td>1.943132e-13</td>\n",
       "      <td>1.941103e-13</td>\n",
       "      <td>9.226339e-16</td>\n",
       "      <td>2.231877e-15</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.050252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>1.044631</td>\n",
       "      <td>7.593942e-07</td>\n",
       "      <td>4.069391e-07</td>\n",
       "      <td>0.000774</td>\n",
       "      <td>0.000755</td>\n",
       "      <td>5.943448e-16</td>\n",
       "      <td>1.871649e-14</td>\n",
       "      <td>2.075432e-15</td>\n",
       "      <td>8.766780e-16</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>753961</td>\n",
       "      <td>1843752</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.042322e-07</td>\n",
       "      <td>6.428994e-07</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>4.764691e-18</td>\n",
       "      <td>3.350644e-20</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>782699</td>\n",
       "      <td>1021154</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>0.812899</td>\n",
       "      <td>1.049173e-06</td>\n",
       "      <td>1.473354e-06</td>\n",
       "      <td>0.000777</td>\n",
       "      <td>0.000784</td>\n",
       "      <td>1.096461e-16</td>\n",
       "      <td>5.206929e-17</td>\n",
       "      <td>5.290043e-16</td>\n",
       "      <td>5.506274e-17</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1752545</td>\n",
       "      <td>1143863</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.288675</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>1.393847</td>\n",
       "      <td>5.178833e-07</td>\n",
       "      <td>2.951272e-07</td>\n",
       "      <td>0.000743</td>\n",
       "      <td>0.000739</td>\n",
       "      <td>2.721570e-19</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>5.106556e-19</td>\n",
       "      <td>4.375358e-19</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       273084           1505602               1                  0   \n",
       "1        38688            353776               1                  0   \n",
       "2       753961           1843752               1                  0   \n",
       "3       782699           1021154               1                  0   \n",
       "4      1752545           1143863               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0                0.0          0.000000               0.0               11   \n",
       "1                0.0          0.050252               0.0               11   \n",
       "2                0.0          0.000000               0.0                1   \n",
       "3                0.0          0.000000               0.0               12   \n",
       "4                0.0          0.288675               0.0                3   \n",
       "\n",
       "   num_followers_d  num_followees_s           ...             weight_f4  \\\n",
       "0                6               15           ...              0.877964   \n",
       "1                6                8           ...              1.044631   \n",
       "2                1                1           ...              3.000000   \n",
       "3               14               12           ...              0.812899   \n",
       "4                2                5           ...              1.393847   \n",
       "\n",
       "    page_rank_s   page_rank_d    katz_s    katz_d        hubs_s        hubs_d  \\\n",
       "0  2.045290e-06  3.459963e-07  0.000773  0.000756  1.943132e-13  1.941103e-13   \n",
       "1  7.593942e-07  4.069391e-07  0.000774  0.000755  5.943448e-16  1.871649e-14   \n",
       "2  4.042322e-07  6.428994e-07  0.000735  0.000735  0.000000e+00  4.764691e-18   \n",
       "3  1.049173e-06  1.473354e-06  0.000777  0.000784  1.096461e-16  5.206929e-17   \n",
       "4  5.178833e-07  2.951272e-07  0.000743  0.000739  2.721570e-19  0.000000e+00   \n",
       "\n",
       "   authorities_s  authorities_d  prefer_Attach_followers  \n",
       "0   9.226339e-16   2.231877e-15                       66  \n",
       "1   2.075432e-15   8.766780e-16                       66  \n",
       "2   3.350644e-20   0.000000e+00                        1  \n",
       "3   5.290043e-16   5.506274e-17                      168  \n",
       "4   5.106556e-19   4.375358e-19                        6  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#for train dataset\n",
    "nfs=np.array(df_final_train['num_followers_s'])\n",
    "nfd=np.array(df_final_train['num_followers_d'])\n",
    "preferential_followers=[]\n",
    "for i in range(len(nfs)):\n",
    "    preferential_followers.append(nfd[i]*nfs[i])\n",
    "df_final_train['prefer_Attach_followers']= preferential_followers\n",
    "df_final_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>weight_f4</th>\n",
       "      <th>page_rank_s</th>\n",
       "      <th>page_rank_d</th>\n",
       "      <th>katz_s</th>\n",
       "      <th>katz_d</th>\n",
       "      <th>hubs_s</th>\n",
       "      <th>hubs_d</th>\n",
       "      <th>authorities_s</th>\n",
       "      <th>authorities_d</th>\n",
       "      <th>prefer_Attach_followers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>848424</td>\n",
       "      <td>784690</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029161</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>1.014128</td>\n",
       "      <td>6.557971e-07</td>\n",
       "      <td>1.559547e-06</td>\n",
       "      <td>0.000754</td>\n",
       "      <td>0.000786</td>\n",
       "      <td>3.243237e-16</td>\n",
       "      <td>1.745627e-16</td>\n",
       "      <td>2.969838e-15</td>\n",
       "      <td>9.269213e-14</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>740824</td>\n",
       "      <td>956372</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.576923</td>\n",
       "      <td>0.144894</td>\n",
       "      <td>0.733674</td>\n",
       "      <td>19</td>\n",
       "      <td>19</td>\n",
       "      <td>22</td>\n",
       "      <td>...</td>\n",
       "      <td>0.640636</td>\n",
       "      <td>5.705371e-07</td>\n",
       "      <td>5.647400e-07</td>\n",
       "      <td>0.000809</td>\n",
       "      <td>0.000809</td>\n",
       "      <td>3.969763e-17</td>\n",
       "      <td>3.961082e-17</td>\n",
       "      <td>3.661442e-18</td>\n",
       "      <td>3.936676e-18</td>\n",
       "      <td>361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1823283</td>\n",
       "      <td>979321</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.109109</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>...</td>\n",
       "      <td>1.048755</td>\n",
       "      <td>1.801494e-06</td>\n",
       "      <td>6.458279e-07</td>\n",
       "      <td>0.000814</td>\n",
       "      <td>0.000739</td>\n",
       "      <td>1.249704e-14</td>\n",
       "      <td>3.305361e-17</td>\n",
       "      <td>1.716478e-14</td>\n",
       "      <td>1.449872e-14</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7964</td>\n",
       "      <td>808296</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>0.781474</td>\n",
       "      <td>6.911796e-07</td>\n",
       "      <td>2.169841e-06</td>\n",
       "      <td>0.000781</td>\n",
       "      <td>0.000828</td>\n",
       "      <td>3.221311e-15</td>\n",
       "      <td>1.257556e-13</td>\n",
       "      <td>3.452734e-14</td>\n",
       "      <td>1.454436e-14</td>\n",
       "      <td>276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1770699</td>\n",
       "      <td>1536004</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.130841</td>\n",
       "      <td>0.081847</td>\n",
       "      <td>0.262336</td>\n",
       "      <td>71</td>\n",
       "      <td>29</td>\n",
       "      <td>89</td>\n",
       "      <td>...</td>\n",
       "      <td>0.393393</td>\n",
       "      <td>1.485611e-06</td>\n",
       "      <td>5.182140e-07</td>\n",
       "      <td>0.001129</td>\n",
       "      <td>0.000893</td>\n",
       "      <td>1.414446e-12</td>\n",
       "      <td>9.676327e-13</td>\n",
       "      <td>7.465039e-13</td>\n",
       "      <td>3.019577e-13</td>\n",
       "      <td>2059</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       848424            784690               1                  0   \n",
       "1       740824            956372               1                  0   \n",
       "2      1823283            979321               1                  0   \n",
       "3         7964            808296               1                  0   \n",
       "4      1770699           1536004               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0           0.000000          0.029161          0.000000                6   \n",
       "1           0.576923          0.144894          0.733674               19   \n",
       "2           0.000000          0.109109          0.000000               21   \n",
       "3           0.000000          0.000000          0.000000               12   \n",
       "4           0.130841          0.081847          0.262336               71   \n",
       "\n",
       "   num_followers_d  num_followees_s           ...             weight_f4  \\\n",
       "0               14                6           ...              1.014128   \n",
       "1               19               22           ...              0.640636   \n",
       "2                2               17           ...              1.048755   \n",
       "3               23               11           ...              0.781474   \n",
       "4               29               89           ...              0.393393   \n",
       "\n",
       "    page_rank_s   page_rank_d    katz_s    katz_d        hubs_s        hubs_d  \\\n",
       "0  6.557971e-07  1.559547e-06  0.000754  0.000786  3.243237e-16  1.745627e-16   \n",
       "1  5.705371e-07  5.647400e-07  0.000809  0.000809  3.969763e-17  3.961082e-17   \n",
       "2  1.801494e-06  6.458279e-07  0.000814  0.000739  1.249704e-14  3.305361e-17   \n",
       "3  6.911796e-07  2.169841e-06  0.000781  0.000828  3.221311e-15  1.257556e-13   \n",
       "4  1.485611e-06  5.182140e-07  0.001129  0.000893  1.414446e-12  9.676327e-13   \n",
       "\n",
       "   authorities_s  authorities_d  prefer_Attach_followers  \n",
       "0   2.969838e-15   9.269213e-14                       84  \n",
       "1   3.661442e-18   3.936676e-18                      361  \n",
       "2   1.716478e-14   1.449872e-14                       42  \n",
       "3   3.452734e-14   1.454436e-14                      276  \n",
       "4   7.465039e-13   3.019577e-13                     2059  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#for test dataset\n",
    "nfs=np.array(df_final_test['num_followers_s'])\n",
    "nfd=np.array(df_final_test['num_followers_d'])\n",
    "preferential_followers=[]\n",
    "for i in range(len(nfs)):\n",
    "    preferential_followers.append(nfd[i]*nfs[i])\n",
    "df_final_test['prefer_Attach_followers']= preferential_followers\n",
    "df_final_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preferential Attachement for followers "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>page_rank_s</th>\n",
       "      <th>page_rank_d</th>\n",
       "      <th>katz_s</th>\n",
       "      <th>katz_d</th>\n",
       "      <th>hubs_s</th>\n",
       "      <th>hubs_d</th>\n",
       "      <th>authorities_s</th>\n",
       "      <th>authorities_d</th>\n",
       "      <th>prefer_Attach_followers</th>\n",
       "      <th>prefer_Attach_followees</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>2.045290e-06</td>\n",
       "      <td>3.459963e-07</td>\n",
       "      <td>0.000773</td>\n",
       "      <td>0.000756</td>\n",
       "      <td>1.943132e-13</td>\n",
       "      <td>1.941103e-13</td>\n",
       "      <td>9.226339e-16</td>\n",
       "      <td>2.231877e-15</td>\n",
       "      <td>66</td>\n",
       "      <td>120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.050252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>7.593942e-07</td>\n",
       "      <td>4.069391e-07</td>\n",
       "      <td>0.000774</td>\n",
       "      <td>0.000755</td>\n",
       "      <td>5.943448e-16</td>\n",
       "      <td>1.871649e-14</td>\n",
       "      <td>2.075432e-15</td>\n",
       "      <td>8.766780e-16</td>\n",
       "      <td>66</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>753961</td>\n",
       "      <td>1843752</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>4.042322e-07</td>\n",
       "      <td>6.428994e-07</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000735</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>4.764691e-18</td>\n",
       "      <td>3.350644e-20</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>782699</td>\n",
       "      <td>1021154</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>1.049173e-06</td>\n",
       "      <td>1.473354e-06</td>\n",
       "      <td>0.000777</td>\n",
       "      <td>0.000784</td>\n",
       "      <td>1.096461e-16</td>\n",
       "      <td>5.206929e-17</td>\n",
       "      <td>5.290043e-16</td>\n",
       "      <td>5.506274e-17</td>\n",
       "      <td>168</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1752545</td>\n",
       "      <td>1143863</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.288675</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>5.178833e-07</td>\n",
       "      <td>2.951272e-07</td>\n",
       "      <td>0.000743</td>\n",
       "      <td>0.000739</td>\n",
       "      <td>2.721570e-19</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>5.106556e-19</td>\n",
       "      <td>4.375358e-19</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       273084           1505602               1                  0   \n",
       "1        38688            353776               1                  0   \n",
       "2       753961           1843752               1                  0   \n",
       "3       782699           1021154               1                  0   \n",
       "4      1752545           1143863               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0                0.0          0.000000               0.0               11   \n",
       "1                0.0          0.050252               0.0               11   \n",
       "2                0.0          0.000000               0.0                1   \n",
       "3                0.0          0.000000               0.0               12   \n",
       "4                0.0          0.288675               0.0                3   \n",
       "\n",
       "   num_followers_d  num_followees_s           ...              page_rank_s  \\\n",
       "0                6               15           ...             2.045290e-06   \n",
       "1                6                8           ...             7.593942e-07   \n",
       "2                1                1           ...             4.042322e-07   \n",
       "3               14               12           ...             1.049173e-06   \n",
       "4                2                5           ...             5.178833e-07   \n",
       "\n",
       "    page_rank_d    katz_s    katz_d        hubs_s        hubs_d  \\\n",
       "0  3.459963e-07  0.000773  0.000756  1.943132e-13  1.941103e-13   \n",
       "1  4.069391e-07  0.000774  0.000755  5.943448e-16  1.871649e-14   \n",
       "2  6.428994e-07  0.000735  0.000735  0.000000e+00  4.764691e-18   \n",
       "3  1.473354e-06  0.000777  0.000784  1.096461e-16  5.206929e-17   \n",
       "4  2.951272e-07  0.000743  0.000739  2.721570e-19  0.000000e+00   \n",
       "\n",
       "   authorities_s  authorities_d  prefer_Attach_followers  \\\n",
       "0   9.226339e-16   2.231877e-15                       66   \n",
       "1   2.075432e-15   8.766780e-16                       66   \n",
       "2   3.350644e-20   0.000000e+00                        1   \n",
       "3   5.290043e-16   5.506274e-17                      168   \n",
       "4   5.106556e-19   4.375358e-19                        6   \n",
       "\n",
       "   prefer_Attach_followees  \n",
       "0                      120  \n",
       "1                       64  \n",
       "2                        2  \n",
       "3                      156  \n",
       "4                        0  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#for train dataset\n",
    "nfs=np.array(df_final_train['num_followees_s'])\n",
    "nfd=np.array(df_final_train['num_followees_d'])\n",
    "preferential_followees=[]\n",
    "for i in range(len(nfs)):\n",
    "    preferential_followees.append(nfd[i]*nfs[i])\n",
    "df_final_train['prefer_Attach_followees']= preferential_followees\n",
    "df_final_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>page_rank_s</th>\n",
       "      <th>page_rank_d</th>\n",
       "      <th>katz_s</th>\n",
       "      <th>katz_d</th>\n",
       "      <th>hubs_s</th>\n",
       "      <th>hubs_d</th>\n",
       "      <th>authorities_s</th>\n",
       "      <th>authorities_d</th>\n",
       "      <th>prefer_Attach_followers</th>\n",
       "      <th>prefer_Attach_followees</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>848424</td>\n",
       "      <td>784690</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029161</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>6.557971e-07</td>\n",
       "      <td>1.559547e-06</td>\n",
       "      <td>0.000754</td>\n",
       "      <td>0.000786</td>\n",
       "      <td>3.243237e-16</td>\n",
       "      <td>1.745627e-16</td>\n",
       "      <td>2.969838e-15</td>\n",
       "      <td>9.269213e-14</td>\n",
       "      <td>84</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>740824</td>\n",
       "      <td>956372</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.576923</td>\n",
       "      <td>0.144894</td>\n",
       "      <td>0.733674</td>\n",
       "      <td>19</td>\n",
       "      <td>19</td>\n",
       "      <td>22</td>\n",
       "      <td>...</td>\n",
       "      <td>5.705371e-07</td>\n",
       "      <td>5.647400e-07</td>\n",
       "      <td>0.000809</td>\n",
       "      <td>0.000809</td>\n",
       "      <td>3.969763e-17</td>\n",
       "      <td>3.961082e-17</td>\n",
       "      <td>3.661442e-18</td>\n",
       "      <td>3.936676e-18</td>\n",
       "      <td>361</td>\n",
       "      <td>418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1823283</td>\n",
       "      <td>979321</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.109109</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>...</td>\n",
       "      <td>1.801494e-06</td>\n",
       "      <td>6.458279e-07</td>\n",
       "      <td>0.000814</td>\n",
       "      <td>0.000739</td>\n",
       "      <td>1.249704e-14</td>\n",
       "      <td>3.305361e-17</td>\n",
       "      <td>1.716478e-14</td>\n",
       "      <td>1.449872e-14</td>\n",
       "      <td>42</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7964</td>\n",
       "      <td>808296</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>6.911796e-07</td>\n",
       "      <td>2.169841e-06</td>\n",
       "      <td>0.000781</td>\n",
       "      <td>0.000828</td>\n",
       "      <td>3.221311e-15</td>\n",
       "      <td>1.257556e-13</td>\n",
       "      <td>3.452734e-14</td>\n",
       "      <td>1.454436e-14</td>\n",
       "      <td>276</td>\n",
       "      <td>242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1770699</td>\n",
       "      <td>1536004</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.130841</td>\n",
       "      <td>0.081847</td>\n",
       "      <td>0.262336</td>\n",
       "      <td>71</td>\n",
       "      <td>29</td>\n",
       "      <td>89</td>\n",
       "      <td>...</td>\n",
       "      <td>1.485611e-06</td>\n",
       "      <td>5.182140e-07</td>\n",
       "      <td>0.001129</td>\n",
       "      <td>0.000893</td>\n",
       "      <td>1.414446e-12</td>\n",
       "      <td>9.676327e-13</td>\n",
       "      <td>7.465039e-13</td>\n",
       "      <td>3.019577e-13</td>\n",
       "      <td>2059</td>\n",
       "      <td>2848</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       848424            784690               1                  0   \n",
       "1       740824            956372               1                  0   \n",
       "2      1823283            979321               1                  0   \n",
       "3         7964            808296               1                  0   \n",
       "4      1770699           1536004               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0           0.000000          0.029161          0.000000                6   \n",
       "1           0.576923          0.144894          0.733674               19   \n",
       "2           0.000000          0.109109          0.000000               21   \n",
       "3           0.000000          0.000000          0.000000               12   \n",
       "4           0.130841          0.081847          0.262336               71   \n",
       "\n",
       "   num_followers_d  num_followees_s           ...              page_rank_s  \\\n",
       "0               14                6           ...             6.557971e-07   \n",
       "1               19               22           ...             5.705371e-07   \n",
       "2                2               17           ...             1.801494e-06   \n",
       "3               23               11           ...             6.911796e-07   \n",
       "4               29               89           ...             1.485611e-06   \n",
       "\n",
       "    page_rank_d    katz_s    katz_d        hubs_s        hubs_d  \\\n",
       "0  1.559547e-06  0.000754  0.000786  3.243237e-16  1.745627e-16   \n",
       "1  5.647400e-07  0.000809  0.000809  3.969763e-17  3.961082e-17   \n",
       "2  6.458279e-07  0.000814  0.000739  1.249704e-14  3.305361e-17   \n",
       "3  2.169841e-06  0.000781  0.000828  3.221311e-15  1.257556e-13   \n",
       "4  5.182140e-07  0.001129  0.000893  1.414446e-12  9.676327e-13   \n",
       "\n",
       "   authorities_s  authorities_d  prefer_Attach_followers  \\\n",
       "0   2.969838e-15   9.269213e-14                       84   \n",
       "1   3.661442e-18   3.936676e-18                      361   \n",
       "2   1.716478e-14   1.449872e-14                       42   \n",
       "3   3.452734e-14   1.454436e-14                      276   \n",
       "4   7.465039e-13   3.019577e-13                     2059   \n",
       "\n",
       "   prefer_Attach_followees  \n",
       "0                       54  \n",
       "1                      418  \n",
       "2                       34  \n",
       "3                      242  \n",
       "4                     2848  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#for test dataset\n",
    "nfs=np.array(df_final_test['num_followees_s'])\n",
    "nfd=np.array(df_final_test['num_followees_d'])\n",
    "preferential_followees=[]\n",
    "for i in range(len(nfs)):\n",
    "    preferential_followees.append(nfd[i]*nfs[i])\n",
    "df_final_test['prefer_Attach_followees']= preferential_followees\n",
    "df_final_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.5 Adding new set of features\n",
    "\n",
    "__we will create these each of these features for both train and test data points__\n",
    "<ol>\n",
    "<li>SVD features for both source and destination</li>\n",
    "</ol>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "def svd(x, S):\n",
    "    try:\n",
    "        z = sadj_dict[x]\n",
    "        return S[z]\n",
    "    except:\n",
    "        return [0,0,0,0,0,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for svd features to get feature vector creating a dict node val and inedx in svd vector\n",
    "sadj_col = sorted(train_graph.nodes())\n",
    "sadj_dict = { val:idx for idx,val in enumerate(sadj_col)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "Adj = nx.adjacency_matrix(train_graph,nodelist=sorted(train_graph.nodes())).asfptype()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adjacency matrix Shape (1780722, 1780722)\n",
      "U Shape (1780722, 6)\n",
      "V Shape (6, 1780722)\n",
      "s Shape (6,)\n"
     ]
    }
   ],
   "source": [
    "U, s, V = svds(Adj, k = 6)\n",
    "print('Adjacency matrix Shape',Adj.shape)\n",
    "print('U Shape',U.shape)\n",
    "print('V Shape',V.shape)\n",
    "print('s Shape',s.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.isfile('data/fea_sample/storage_sample_stage4.h5'):\n",
    "    #===================================================================================================\n",
    "    \n",
    "    df_final_train[['svd_u_s_1', 'svd_u_s_2','svd_u_s_3', 'svd_u_s_4', 'svd_u_s_5', 'svd_u_s_6']] = \\\n",
    "    df_final_train.source_node.apply(lambda x: svd(x, U)).apply(pd.Series)\n",
    "    \n",
    "    df_final_train[['svd_u_d_1', 'svd_u_d_2', 'svd_u_d_3', 'svd_u_d_4', 'svd_u_d_5','svd_u_d_6']] = \\\n",
    "    df_final_train.destination_node.apply(lambda x: svd(x, U)).apply(pd.Series)\n",
    "    #===================================================================================================\n",
    "    \n",
    "    df_final_train[['svd_v_s_1','svd_v_s_2', 'svd_v_s_3', 'svd_v_s_4', 'svd_v_s_5', 'svd_v_s_6',]] = \\\n",
    "    df_final_train.source_node.apply(lambda x: svd(x, V.T)).apply(pd.Series)\n",
    "\n",
    "    df_final_train[['svd_v_d_1', 'svd_v_d_2', 'svd_v_d_3', 'svd_v_d_4', 'svd_v_d_5','svd_v_d_6']] = \\\n",
    "    df_final_train.destination_node.apply(lambda x: svd(x, V.T)).apply(pd.Series)\n",
    "    #===================================================================================================\n",
    "    \n",
    "    df_final_test[['svd_u_s_1', 'svd_u_s_2','svd_u_s_3', 'svd_u_s_4', 'svd_u_s_5', 'svd_u_s_6']] = \\\n",
    "    df_final_test.source_node.apply(lambda x: svd(x, U)).apply(pd.Series)\n",
    "    \n",
    "    df_final_test[['svd_u_d_1', 'svd_u_d_2', 'svd_u_d_3', 'svd_u_d_4', 'svd_u_d_5','svd_u_d_6']] = \\\n",
    "    df_final_test.destination_node.apply(lambda x: svd(x, U)).apply(pd.Series)\n",
    "\n",
    "    #===================================================================================================\n",
    "    \n",
    "    df_final_test[['svd_v_s_1','svd_v_s_2', 'svd_v_s_3', 'svd_v_s_4', 'svd_v_s_5', 'svd_v_s_6',]] = \\\n",
    "    df_final_test.source_node.apply(lambda x: svd(x, V.T)).apply(pd.Series)\n",
    "\n",
    "    df_final_test[['svd_v_d_1', 'svd_v_d_2', 'svd_v_d_3', 'svd_v_d_4', 'svd_v_d_5','svd_v_d_6']] = \\\n",
    "    df_final_test.destination_node.apply(lambda x: svd(x, V.T)).apply(pd.Series)\n",
    "    #===================================================================================================\n",
    "\n",
    "    hdf = HDFStore('data/fea_sample/storage_sample_stage4.h5')\n",
    "    hdf.put('train_df',df_final_train, format='table', data_columns=True)\n",
    "    hdf.put('test_df',df_final_test, format='table', data_columns=True)\n",
    "    hdf.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100002, 57)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>svd_v_s_3</th>\n",
       "      <th>svd_v_s_4</th>\n",
       "      <th>svd_v_s_5</th>\n",
       "      <th>svd_v_s_6</th>\n",
       "      <th>svd_v_d_1</th>\n",
       "      <th>svd_v_d_2</th>\n",
       "      <th>svd_v_d_3</th>\n",
       "      <th>svd_v_d_4</th>\n",
       "      <th>svd_v_d_5</th>\n",
       "      <th>svd_v_d_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>1.983705e-06</td>\n",
       "      <td>1.545076e-13</td>\n",
       "      <td>8.108423e-13</td>\n",
       "      <td>1.719704e-14</td>\n",
       "      <td>-1.355368e-12</td>\n",
       "      <td>4.675301e-13</td>\n",
       "      <td>1.128590e-06</td>\n",
       "      <td>6.616666e-14</td>\n",
       "      <td>9.771076e-13</td>\n",
       "      <td>4.160020e-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.050252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>1.705842e-09</td>\n",
       "      <td>3.701182e-13</td>\n",
       "      <td>1.613518e-12</td>\n",
       "      <td>3.868404e-14</td>\n",
       "      <td>-8.625225e-10</td>\n",
       "      <td>3.900850e-11</td>\n",
       "      <td>1.299678e-11</td>\n",
       "      <td>8.522088e-14</td>\n",
       "      <td>2.913150e-13</td>\n",
       "      <td>1.634056e-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>753961</td>\n",
       "      <td>1843752</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>3.823744e-16</td>\n",
       "      <td>1.880190e-19</td>\n",
       "      <td>4.484727e-19</td>\n",
       "      <td>6.201571e-19</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>782699</td>\n",
       "      <td>1021154</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>4.099766e-11</td>\n",
       "      <td>2.621501e-13</td>\n",
       "      <td>6.220152e-12</td>\n",
       "      <td>9.860256e-15</td>\n",
       "      <td>-4.980536e-13</td>\n",
       "      <td>6.743302e-10</td>\n",
       "      <td>7.931299e-12</td>\n",
       "      <td>8.001806e-13</td>\n",
       "      <td>5.325364e-11</td>\n",
       "      <td>1.026230e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1752545</td>\n",
       "      <td>1143863</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.288675</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>7.849315e-13</td>\n",
       "      <td>3.046905e-15</td>\n",
       "      <td>4.786346e-15</td>\n",
       "      <td>9.508786e-18</td>\n",
       "      <td>-1.035927e-15</td>\n",
       "      <td>1.007523e-13</td>\n",
       "      <td>7.845564e-13</td>\n",
       "      <td>2.910923e-15</td>\n",
       "      <td>4.769340e-15</td>\n",
       "      <td>8.147576e-18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 57 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       273084           1505602               1                  0   \n",
       "1        38688            353776               1                  0   \n",
       "2       753961           1843752               1                  0   \n",
       "3       782699           1021154               1                  0   \n",
       "4      1752545           1143863               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0                0.0          0.000000               0.0               11   \n",
       "1                0.0          0.050252               0.0               11   \n",
       "2                0.0          0.000000               0.0                1   \n",
       "3                0.0          0.000000               0.0               12   \n",
       "4                0.0          0.288675               0.0                3   \n",
       "\n",
       "   num_followers_d  num_followees_s      ...          svd_v_s_3     svd_v_s_4  \\\n",
       "0                6               15      ...       1.983705e-06  1.545076e-13   \n",
       "1                6                8      ...       1.705842e-09  3.701182e-13   \n",
       "2                1                1      ...       3.823744e-16  1.880190e-19   \n",
       "3               14               12      ...       4.099766e-11  2.621501e-13   \n",
       "4                2                5      ...       7.849315e-13  3.046905e-15   \n",
       "\n",
       "      svd_v_s_5     svd_v_s_6     svd_v_d_1     svd_v_d_2     svd_v_d_3  \\\n",
       "0  8.108423e-13  1.719704e-14 -1.355368e-12  4.675301e-13  1.128590e-06   \n",
       "1  1.613518e-12  3.868404e-14 -8.625225e-10  3.900850e-11  1.299678e-11   \n",
       "2  4.484727e-19  6.201571e-19  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "3  6.220152e-12  9.860256e-15 -4.980536e-13  6.743302e-10  7.931299e-12   \n",
       "4  4.786346e-15  9.508786e-18 -1.035927e-15  1.007523e-13  7.845564e-13   \n",
       "\n",
       "      svd_v_d_4     svd_v_d_5     svd_v_d_6  \n",
       "0  6.616666e-14  9.771076e-13  4.160020e-14  \n",
       "1  8.522088e-14  2.913150e-13  1.634056e-14  \n",
       "2  0.000000e+00  0.000000e+00  0.000000e+00  \n",
       "3  8.001806e-13  5.325364e-11  1.026230e-15  \n",
       "4  2.910923e-15  4.769340e-15  8.147576e-18  \n",
       "\n",
       "[5 rows x 57 columns]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['source_node', 'destination_node', 'indicator_link',\n",
       "       'jaccard_followers', 'jaccard_followees', 'cosine_followers',\n",
       "       'cosine_followees', 'num_followers_s', 'num_followers_d',\n",
       "       'num_followees_s', 'num_followees_d', 'inter_followers',\n",
       "       'inter_followees', 'adar_index', 'follows_back', 'same_comp',\n",
       "       'shortest_path', 'weight_in', 'weight_out', 'weight_f1', 'weight_f2',\n",
       "       'weight_f3', 'weight_f4', 'page_rank_s', 'page_rank_d', 'katz_s',\n",
       "       'katz_d', 'hubs_s', 'hubs_d', 'authorities_s', 'authorities_d',\n",
       "       'prefer_Attach_followers', 'prefer_Attach_followees', 'svd_u_s_1',\n",
       "       'svd_u_s_2', 'svd_u_s_3', 'svd_u_s_4', 'svd_u_s_5', 'svd_u_s_6',\n",
       "       'svd_u_d_1', 'svd_u_d_2', 'svd_u_d_3', 'svd_u_d_4', 'svd_u_d_5',\n",
       "       'svd_u_d_6', 'svd_v_s_1', 'svd_v_s_2', 'svd_v_s_3', 'svd_v_s_4',\n",
       "       'svd_v_s_5', 'svd_v_s_6', 'svd_v_d_1', 'svd_v_d_2', 'svd_v_d_3',\n",
       "       'svd_v_d_4', 'svd_v_d_5', 'svd_v_d_6'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adding feature svd_dot "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "svd_dot is Dot product between sourse node svd and destination node svd features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for train datasets\n",
    "s1,s2,s3,s4,s5,s6=df_final_train['svd_u_s_1'],df_final_train['svd_u_s_2'],df_final_train['svd_u_s_3'],df_final_train['svd_u_s_4'],df_final_train['svd_u_s_5'],df_final_train['svd_u_s_6']\n",
    "s7,s8,s9,s10,s11,s12=df_final_train['svd_v_s_1'],df_final_train['svd_v_s_2'],df_final_train['svd_v_s_3'],df_final_train['svd_v_s_4'],df_final_train['svd_v_s_5'],df_final_train['svd_v_s_6']\n",
    "\n",
    "d1,d2,d3,d4,d5,d6=df_final_train['svd_u_d_1'],df_final_train['svd_u_d_2'],df_final_train['svd_u_d_3'],df_final_train['svd_u_d_4'],df_final_train['svd_u_d_5'],df_final_train['svd_u_d_6']\n",
    "d7,d8,d9,d10,d11,d12=df_final_train['svd_v_d_1'],df_final_train['svd_v_d_2'],df_final_train['svd_v_d_3'],df_final_train['svd_v_d_4'],df_final_train['svd_v_d_5'],df_final_train['svd_v_d_6']\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "svd_dot_u=[]\n",
    "for i in range(len(np.array(s1))):\n",
    "    a=[]\n",
    "    b=[]\n",
    "\n",
    "    a.append(np.array(s1[i]))\n",
    "    a.append(np.array(s2[i]))\n",
    "    a.append(np.array(s3[i]))\n",
    "    a.append(np.array(s4[i]))\n",
    "    a.append(np.array(s5[i]))\n",
    "    a.append(np.array(s6[i]))\n",
    "    \n",
    "    b.append(np.array(d1[i]))\n",
    "    b.append(np.array(d2[i]))\n",
    "    b.append(np.array(d3[i]))\n",
    "    b.append(np.array(d4[i]))\n",
    "    b.append(np.array(d5[i]))\n",
    "    b.append(np.array(d6[i]))\n",
    "    \n",
    "    svd_dot_u.append(np.dot(a,b))\n",
    "df_final_train['svd_dot_u']=svd_dot_u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "svd_dot_v=[]\n",
    "for i in range(len(np.array(s7))):\n",
    "    c=[]\n",
    "    d=[]\n",
    "    \n",
    "    c.append(np.array(s7[i]))\n",
    "    c.append(np.array(s8[i]))\n",
    "    c.append(np.array(s9[i]))\n",
    "    c.append(np.array(s10[i]))\n",
    "    c.append(np.array(s11[i]))\n",
    "    c.append(np.array(s12[i]))\n",
    "   \n",
    "    d.append(np.array(d7[i]))\n",
    "    d.append(np.array(d8[i]))\n",
    "    d.append(np.array(d9[i]))\n",
    "    d.append(np.array(d10[i]))\n",
    "    d.append(np.array(d11[i]))\n",
    "    d.append(np.array(d12[i]))\n",
    "    svd_dot_v.append(np.dot(c,d))\n",
    "    \n",
    "df_final_train['svd_dot_v']=svd_dot_v"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>svd_v_s_5</th>\n",
       "      <th>svd_v_s_6</th>\n",
       "      <th>svd_v_d_1</th>\n",
       "      <th>svd_v_d_2</th>\n",
       "      <th>svd_v_d_3</th>\n",
       "      <th>svd_v_d_4</th>\n",
       "      <th>svd_v_d_5</th>\n",
       "      <th>svd_v_d_6</th>\n",
       "      <th>svd_dot_u</th>\n",
       "      <th>svd_dot_v</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>273084</td>\n",
       "      <td>1505602</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>8.108423e-13</td>\n",
       "      <td>1.719704e-14</td>\n",
       "      <td>-1.355368e-12</td>\n",
       "      <td>4.675301e-13</td>\n",
       "      <td>1.128590e-06</td>\n",
       "      <td>6.616666e-14</td>\n",
       "      <td>9.771076e-13</td>\n",
       "      <td>4.160020e-14</td>\n",
       "      <td>1.114944e-11</td>\n",
       "      <td>2.238790e-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38688</td>\n",
       "      <td>353776</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.050252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>1.613518e-12</td>\n",
       "      <td>3.868404e-14</td>\n",
       "      <td>-8.625225e-10</td>\n",
       "      <td>3.900850e-11</td>\n",
       "      <td>1.299678e-11</td>\n",
       "      <td>8.522088e-14</td>\n",
       "      <td>2.913150e-13</td>\n",
       "      <td>1.634056e-14</td>\n",
       "      <td>3.166610e-19</td>\n",
       "      <td>8.691193e-19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>753961</td>\n",
       "      <td>1843752</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>4.484727e-19</td>\n",
       "      <td>6.201571e-19</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>4.830463e-33</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>782699</td>\n",
       "      <td>1021154</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>6.220152e-12</td>\n",
       "      <td>9.860256e-15</td>\n",
       "      <td>-4.980536e-13</td>\n",
       "      <td>6.743302e-10</td>\n",
       "      <td>7.931299e-12</td>\n",
       "      <td>8.001806e-13</td>\n",
       "      <td>5.325364e-11</td>\n",
       "      <td>1.026230e-15</td>\n",
       "      <td>8.515352e-24</td>\n",
       "      <td>1.215102e-21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1752545</td>\n",
       "      <td>1143863</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.288675</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4.786346e-15</td>\n",
       "      <td>9.508786e-18</td>\n",
       "      <td>-1.035927e-15</td>\n",
       "      <td>1.007523e-13</td>\n",
       "      <td>7.845564e-13</td>\n",
       "      <td>2.910923e-15</td>\n",
       "      <td>4.769340e-15</td>\n",
       "      <td>8.147576e-18</td>\n",
       "      <td>1.029774e-31</td>\n",
       "      <td>6.203340e-25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 59 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       273084           1505602               1                  0   \n",
       "1        38688            353776               1                  0   \n",
       "2       753961           1843752               1                  0   \n",
       "3       782699           1021154               1                  0   \n",
       "4      1752545           1143863               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0                0.0          0.000000               0.0               11   \n",
       "1                0.0          0.050252               0.0               11   \n",
       "2                0.0          0.000000               0.0                1   \n",
       "3                0.0          0.000000               0.0               12   \n",
       "4                0.0          0.288675               0.0                3   \n",
       "\n",
       "   num_followers_d  num_followees_s      ...          svd_v_s_5     svd_v_s_6  \\\n",
       "0                6               15      ...       8.108423e-13  1.719704e-14   \n",
       "1                6                8      ...       1.613518e-12  3.868404e-14   \n",
       "2                1                1      ...       4.484727e-19  6.201571e-19   \n",
       "3               14               12      ...       6.220152e-12  9.860256e-15   \n",
       "4                2                5      ...       4.786346e-15  9.508786e-18   \n",
       "\n",
       "      svd_v_d_1     svd_v_d_2     svd_v_d_3     svd_v_d_4     svd_v_d_5  \\\n",
       "0 -1.355368e-12  4.675301e-13  1.128590e-06  6.616666e-14  9.771076e-13   \n",
       "1 -8.625225e-10  3.900850e-11  1.299678e-11  8.522088e-14  2.913150e-13   \n",
       "2  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "3 -4.980536e-13  6.743302e-10  7.931299e-12  8.001806e-13  5.325364e-11   \n",
       "4 -1.035927e-15  1.007523e-13  7.845564e-13  2.910923e-15  4.769340e-15   \n",
       "\n",
       "      svd_v_d_6     svd_dot_u     svd_dot_v  \n",
       "0  4.160020e-14  1.114944e-11  2.238790e-12  \n",
       "1  1.634056e-14  3.166610e-19  8.691193e-19  \n",
       "2  0.000000e+00  4.830463e-33  0.000000e+00  \n",
       "3  1.026230e-15  8.515352e-24  1.215102e-21  \n",
       "4  8.147576e-18  1.029774e-31  6.203340e-25  \n",
       "\n",
       "[5 rows x 59 columns]"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for test dataset\n",
    "s1,s2,s3,s4,s5,s6=df_final_test['svd_u_s_1'],df_final_test['svd_u_s_2'],df_final_test['svd_u_s_3'],df_final_test['svd_u_s_4'],df_final_test['svd_u_s_5'],df_final_test['svd_u_s_6']\n",
    "s7,s8,s9,s10,s11,s12=df_final_test['svd_v_s_1'],df_final_test['svd_v_s_2'],df_final_test['svd_v_s_3'],df_final_test['svd_v_s_4'],df_final_test['svd_v_s_5'],df_final_test['svd_v_s_6']\n",
    "\n",
    "d1,d2,d3,d4,d5,d6=df_final_test['svd_u_d_1'],df_final_test['svd_u_d_2'],df_final_test['svd_u_d_3'],df_final_test['svd_u_d_4'],df_final_test['svd_u_d_5'],df_final_test['svd_u_d_6']\n",
    "d7,d8,d9,d10,d11,d12=df_final_test['svd_v_d_1'],df_final_test['svd_v_d_2'],df_final_test['svd_v_d_3'],df_final_test['svd_v_d_4'],df_final_test['svd_v_d_5'],df_final_test['svd_v_d_6']\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "svd_dot_u=[]\n",
    "for i in range(len(np.array(s1))):\n",
    "    a=[]\n",
    "    b=[]\n",
    "\n",
    "    a.append(np.array(s1[i]))\n",
    "    a.append(np.array(s2[i]))\n",
    "    a.append(np.array(s3[i]))\n",
    "    a.append(np.array(s4[i]))\n",
    "    a.append(np.array(s5[i]))\n",
    "    a.append(np.array(s6[i]))\n",
    "    \n",
    "    b.append(np.array(d1[i]))\n",
    "    b.append(np.array(d2[i]))\n",
    "    b.append(np.array(d3[i]))\n",
    "    b.append(np.array(d4[i]))\n",
    "    b.append(np.array(d5[i]))\n",
    "    b.append(np.array(d6[i]))\n",
    "    \n",
    "    svd_dot_u.append(np.dot(a,b))\n",
    "df_final_test['svd_dot_u']=svd_dot_u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "svd_dot_v=[]\n",
    "for i in range(len(np.array(s7))):\n",
    "    c=[]\n",
    "    d=[]\n",
    "    \n",
    "    c.append(np.array(s7[i]))\n",
    "    c.append(np.array(s8[i]))\n",
    "    c.append(np.array(s9[i]))\n",
    "    c.append(np.array(s10[i]))\n",
    "    c.append(np.array(s11[i]))\n",
    "    c.append(np.array(s12[i]))\n",
    "   \n",
    "    d.append(np.array(d7[i]))\n",
    "    d.append(np.array(d8[i]))\n",
    "    d.append(np.array(d9[i]))\n",
    "    d.append(np.array(d10[i]))\n",
    "    d.append(np.array(d11[i]))\n",
    "    d.append(np.array(d12[i]))\n",
    "    svd_dot_v.append(np.dot(c,d))\n",
    "    \n",
    "df_final_test['svd_dot_v']=svd_dot_v"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>source_node</th>\n",
       "      <th>destination_node</th>\n",
       "      <th>indicator_link</th>\n",
       "      <th>jaccard_followers</th>\n",
       "      <th>jaccard_followees</th>\n",
       "      <th>cosine_followers</th>\n",
       "      <th>cosine_followees</th>\n",
       "      <th>num_followers_s</th>\n",
       "      <th>num_followers_d</th>\n",
       "      <th>num_followees_s</th>\n",
       "      <th>...</th>\n",
       "      <th>svd_v_s_5</th>\n",
       "      <th>svd_v_s_6</th>\n",
       "      <th>svd_v_d_1</th>\n",
       "      <th>svd_v_d_2</th>\n",
       "      <th>svd_v_d_3</th>\n",
       "      <th>svd_v_d_4</th>\n",
       "      <th>svd_v_d_5</th>\n",
       "      <th>svd_v_d_6</th>\n",
       "      <th>svd_dot_u</th>\n",
       "      <th>svd_dot_v</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>848424</td>\n",
       "      <td>784690</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029161</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>4.341619e-13</td>\n",
       "      <td>5.535500e-14</td>\n",
       "      <td>-9.994077e-10</td>\n",
       "      <td>5.791914e-10</td>\n",
       "      <td>3.512361e-07</td>\n",
       "      <td>2.486660e-09</td>\n",
       "      <td>2.771146e-09</td>\n",
       "      <td>1.727695e-12</td>\n",
       "      <td>8.425165e-20</td>\n",
       "      <td>2.074816e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>740824</td>\n",
       "      <td>956372</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.576923</td>\n",
       "      <td>0.144894</td>\n",
       "      <td>0.733674</td>\n",
       "      <td>19</td>\n",
       "      <td>19</td>\n",
       "      <td>22</td>\n",
       "      <td>...</td>\n",
       "      <td>4.009029e-17</td>\n",
       "      <td>6.765513e-17</td>\n",
       "      <td>3.298228e-19</td>\n",
       "      <td>3.587568e-16</td>\n",
       "      <td>3.016322e-14</td>\n",
       "      <td>1.027750e-14</td>\n",
       "      <td>4.312828e-17</td>\n",
       "      <td>7.292144e-17</td>\n",
       "      <td>5.764647e-27</td>\n",
       "      <td>9.456294e-28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1823283</td>\n",
       "      <td>979321</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.109109</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>...</td>\n",
       "      <td>5.436900e-11</td>\n",
       "      <td>3.199352e-13</td>\n",
       "      <td>-8.331557e-06</td>\n",
       "      <td>2.374065e-10</td>\n",
       "      <td>2.456697e-10</td>\n",
       "      <td>6.434954e-12</td>\n",
       "      <td>4.163228e-11</td>\n",
       "      <td>2.702427e-13</td>\n",
       "      <td>8.583387e-20</td>\n",
       "      <td>6.955785e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7964</td>\n",
       "      <td>808296</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>2.629888e-09</td>\n",
       "      <td>6.435541e-13</td>\n",
       "      <td>-8.947553e-12</td>\n",
       "      <td>2.430581e-12</td>\n",
       "      <td>2.612161e-05</td>\n",
       "      <td>2.023465e-12</td>\n",
       "      <td>3.669848e-12</td>\n",
       "      <td>2.710928e-13</td>\n",
       "      <td>1.465291e-09</td>\n",
       "      <td>4.032480e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1770699</td>\n",
       "      <td>1536004</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.130841</td>\n",
       "      <td>0.081847</td>\n",
       "      <td>0.262336</td>\n",
       "      <td>71</td>\n",
       "      <td>29</td>\n",
       "      <td>89</td>\n",
       "      <td>...</td>\n",
       "      <td>1.873245e-10</td>\n",
       "      <td>1.391411e-11</td>\n",
       "      <td>5.949877e-11</td>\n",
       "      <td>-4.555330e-11</td>\n",
       "      <td>4.615655e-03</td>\n",
       "      <td>2.121942e-11</td>\n",
       "      <td>4.955822e-11</td>\n",
       "      <td>5.628206e-12</td>\n",
       "      <td>4.698788e-04</td>\n",
       "      <td>8.573668e-05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 59 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   source_node  destination_node  indicator_link  jaccard_followers  \\\n",
       "0       848424            784690               1                  0   \n",
       "1       740824            956372               1                  0   \n",
       "2      1823283            979321               1                  0   \n",
       "3         7964            808296               1                  0   \n",
       "4      1770699           1536004               1                  0   \n",
       "\n",
       "   jaccard_followees  cosine_followers  cosine_followees  num_followers_s  \\\n",
       "0           0.000000          0.029161          0.000000                6   \n",
       "1           0.576923          0.144894          0.733674               19   \n",
       "2           0.000000          0.109109          0.000000               21   \n",
       "3           0.000000          0.000000          0.000000               12   \n",
       "4           0.130841          0.081847          0.262336               71   \n",
       "\n",
       "   num_followers_d  num_followees_s      ...          svd_v_s_5     svd_v_s_6  \\\n",
       "0               14                6      ...       4.341619e-13  5.535500e-14   \n",
       "1               19               22      ...       4.009029e-17  6.765513e-17   \n",
       "2                2               17      ...       5.436900e-11  3.199352e-13   \n",
       "3               23               11      ...       2.629888e-09  6.435541e-13   \n",
       "4               29               89      ...       1.873245e-10  1.391411e-11   \n",
       "\n",
       "      svd_v_d_1     svd_v_d_2     svd_v_d_3     svd_v_d_4     svd_v_d_5  \\\n",
       "0 -9.994077e-10  5.791914e-10  3.512361e-07  2.486660e-09  2.771146e-09   \n",
       "1  3.298228e-19  3.587568e-16  3.016322e-14  1.027750e-14  4.312828e-17   \n",
       "2 -8.331557e-06  2.374065e-10  2.456697e-10  6.434954e-12  4.163228e-11   \n",
       "3 -8.947553e-12  2.430581e-12  2.612161e-05  2.023465e-12  3.669848e-12   \n",
       "4  5.949877e-11 -4.555330e-11  4.615655e-03  2.121942e-11  4.955822e-11   \n",
       "\n",
       "      svd_v_d_6     svd_dot_u     svd_dot_v  \n",
       "0  1.727695e-12  8.425165e-20  2.074816e-17  \n",
       "1  7.292144e-17  5.764647e-27  9.456294e-28  \n",
       "2  2.702427e-13  8.583387e-20  6.955785e-11  \n",
       "3  2.710928e-13  1.465291e-09  4.032480e-10  \n",
       "4  5.628206e-12  4.698788e-04  8.573668e-05  \n",
       "\n",
       "[5 rows x 59 columns]"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "hdf = HDFStore('data/fea_sample/storage_sample_stage5.h5')\n",
    "hdf.put('train_df',df_final_train, format='table', data_columns=True)\n",
    "hdf.put('test_df',df_final_test, format='table', data_columns=True)\n",
    "hdf.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandas import HDFStore,DataFrame\n",
    "from pandas import read_hdf\n",
    "from scipy.sparse.linalg import svds, eigs\n",
    "import gc\n",
    "from tqdm import tqdm\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100002, 59)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Index(['source_node', 'destination_node', 'indicator_link',\n",
       "       'jaccard_followers', 'jaccard_followees', 'cosine_followers',\n",
       "       'cosine_followees', 'num_followers_s', 'num_followers_d',\n",
       "       'num_followees_s', 'num_followees_d', 'inter_followers',\n",
       "       'inter_followees', 'adar_index', 'follows_back', 'same_comp',\n",
       "       'shortest_path', 'weight_in', 'weight_out', 'weight_f1', 'weight_f2',\n",
       "       'weight_f3', 'weight_f4', 'page_rank_s', 'page_rank_d', 'katz_s',\n",
       "       'katz_d', 'hubs_s', 'hubs_d', 'authorities_s', 'authorities_d',\n",
       "       'prefer_Attach_followers', 'prefer_Attach_followees', 'svd_u_s_1',\n",
       "       'svd_u_s_2', 'svd_u_s_3', 'svd_u_s_4', 'svd_u_s_5', 'svd_u_s_6',\n",
       "       'svd_u_d_1', 'svd_u_d_2', 'svd_u_d_3', 'svd_u_d_4', 'svd_u_d_5',\n",
       "       'svd_u_d_6', 'svd_v_s_1', 'svd_v_s_2', 'svd_v_s_3', 'svd_v_s_4',\n",
       "       'svd_v_s_5', 'svd_v_s_6', 'svd_v_d_1', 'svd_v_d_2', 'svd_v_d_3',\n",
       "       'svd_v_d_4', 'svd_v_d_5', 'svd_v_d_6', 'svd_dot_u', 'svd_dot_v'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df_final_train.shape)\n",
    "df_final_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = df_final_train.indicator_link\n",
    "y_test = df_final_test.indicator_link"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final_train.drop(['source_node', 'destination_node','indicator_link'],axis=1,inplace=True)\n",
    "df_final_test.drop(['source_node', 'destination_node','indicator_link'],axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100002, 56)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Building Random Forest model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimators =  10 Train Score 0.9045367146494829 test Score 0.8733556539583108\n",
      "Estimators =  50 Train Score 0.9216537231612608 test Score 0.9099085721558727\n",
      "Estimators =  100 Train Score 0.922092975121083 test Score 0.9148909289985682\n",
      "Estimators =  250 Train Score 0.9216755098656939 test Score 0.912270327349525\n",
      "Estimators =  450 Train Score 0.9222757156908422 test Score 0.915440479949479\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Estimators vs score at depth of 5')"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "estimators = [10,50,100,250,450]\n",
    "train_scores = []\n",
    "test_scores = []\n",
    "for i in estimators:\n",
    "    clf = RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
    "            max_depth=5, max_features='auto', max_leaf_nodes=None,\n",
    "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "            min_samples_leaf=52, min_samples_split=120,\n",
    "            min_weight_fraction_leaf=0.0, n_estimators=i, n_jobs=-1,random_state=25,verbose=0,warm_start=False)\n",
    "    clf.fit(df_final_train,y_train)\n",
    "    train_sc = f1_score(y_train,clf.predict(df_final_train))\n",
    "    test_sc = f1_score(y_test,clf.predict(df_final_test))\n",
    "    test_scores.append(test_sc)\n",
    "    train_scores.append(train_sc)\n",
    "    print('Estimators = ',i,'Train Score',train_sc,'test Score',test_sc)\n",
    "plt.plot(estimators,train_scores,label='Train Score')\n",
    "plt.plot(estimators,test_scores,label='Test Score')\n",
    "plt.xlabel('Estimators')\n",
    "plt.ylabel('Score')\n",
    "plt.title('Estimators vs score at depth of 5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "depth =  3 Train Score 0.8849736187572197 test Score 0.860825849697074\n",
      "depth =  9 Train Score 0.9593731041456016 test Score 0.922647021465769\n",
      "depth =  11 Train Score 0.9612595439095484 test Score 0.922985857133797\n",
      "depth =  15 Train Score 0.963692904163806 test Score 0.9246409963146526\n",
      "depth =  20 Train Score 0.9642366997658836 test Score 0.9236508542117088\n",
      "depth =  35 Train Score 0.9643120200990829 test Score 0.9190983972926376\n",
      "depth =  50 Train Score 0.9643120200990829 test Score 0.9190983972926376\n",
      "depth =  70 Train Score 0.9643120200990829 test Score 0.9190983972926376\n",
      "depth =  130 Train Score 0.9643120200990829 test Score 0.9190983972926376\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "depths = [3,9,11,15,20,35,50,70,130]\n",
    "train_scores = []\n",
    "test_scores = []\n",
    "for i in depths:\n",
    "    clf = RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
    "            max_depth=i, max_features='auto', max_leaf_nodes=None,\n",
    "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "            min_samples_leaf=52, min_samples_split=120,\n",
    "            min_weight_fraction_leaf=0.0, n_estimators=115, n_jobs=-1,random_state=25,verbose=0,warm_start=False)\n",
    "    clf.fit(df_final_train,y_train)\n",
    "    train_sc = f1_score(y_train,clf.predict(df_final_train))\n",
    "    test_sc = f1_score(y_test,clf.predict(df_final_test))\n",
    "    test_scores.append(test_sc)\n",
    "    train_scores.append(train_sc)\n",
    "    print('depth = ',i,'Train Score',train_sc,'test Score',test_sc)\n",
    "plt.plot(depths,train_scores,label='Train Score')\n",
    "plt.plot(depths,test_scores,label='Test Score')\n",
    "plt.xlabel('Depth')\n",
    "plt.ylabel('Score')\n",
    "plt.title('Depth vs score at depth of 5 at estimators = 115')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean test scores [0.96264378 0.96199548 0.96018836 0.96193459 0.96359382]\n",
      "mean train scores [0.96367319 0.96299694 0.96089724 0.96305652 0.96520757]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from scipy.stats import randint as sp_randint\n",
    "from scipy.stats import uniform\n",
    "\n",
    "param_dist = {\"n_estimators\":sp_randint(105,125),\n",
    "              \"max_depth\": sp_randint(10,15),\n",
    "              \"min_samples_split\": sp_randint(110,190),\n",
    "              \"min_samples_leaf\": sp_randint(25,65)}\n",
    "\n",
    "clf = RandomForestClassifier(random_state=25,n_jobs=-1)\n",
    "\n",
    "rf_random = RandomizedSearchCV(clf, param_distributions=param_dist,\n",
    "                                   n_iter=5,cv=10,scoring='f1',random_state=25)\n",
    "\n",
    "rf_random.fit(df_final_train,y_train)\n",
    "print('mean test scores',rf_random.cv_results_['mean_test_score'])\n",
    "print('mean train scores',rf_random.cv_results_['mean_train_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "            max_depth=14, max_features='auto', max_leaf_nodes=None,\n",
      "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "            min_samples_leaf=28, min_samples_split=111,\n",
      "            min_weight_fraction_leaf=0.0, n_estimators=121, n_jobs=-1,\n",
      "            oob_score=False, random_state=25, verbose=0, warm_start=False)\n"
     ]
    }
   ],
   "source": [
    "print(rf_random.best_estimator_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
    "            max_depth=14, max_features='auto', max_leaf_nodes=None,\n",
    "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "            min_samples_leaf=28, min_samples_split=111,\n",
    "            min_weight_fraction_leaf=0.0, n_estimators=121, n_jobs=-1,\n",
    "            oob_score=False, random_state=25, verbose=0, warm_start=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf.fit(df_final_train,y_train)\n",
    "y_train_pred = clf.predict(df_final_train)\n",
    "y_test_pred = clf.predict(df_final_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train f1 score 0.964842882802882\n",
      "Test f1 score 0.9149427252521798\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "print('Train f1 score',f1_score(y_train,y_train_pred))\n",
    "print('Test f1 score',f1_score(y_test,y_test_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "def plot_confusion_matrix(test_y, predict_y):\n",
    "    C = confusion_matrix(test_y, predict_y)\n",
    "    \n",
    "    A =(((C.T)/(C.sum(axis=1))).T)\n",
    "    \n",
    "    B =(C/C.sum(axis=0))\n",
    "    plt.figure(figsize=(20,4))\n",
    "    \n",
    "    labels = [0,1]\n",
    "    # representing A in heatmap format\n",
    "    cmap=sns.light_palette(\"blue\")\n",
    "    plt.subplot(1, 3, 1)\n",
    "    sns.heatmap(C, annot=True, cmap=cmap, fmt=\".3f\", xticklabels=labels, yticklabels=labels)\n",
    "    plt.xlabel('Predicted Class')\n",
    "    plt.ylabel('Original Class')\n",
    "    plt.title(\"Confusion matrix\")\n",
    "    \n",
    "    plt.subplot(1, 3, 2)\n",
    "    sns.heatmap(B, annot=True, cmap=cmap, fmt=\".3f\", xticklabels=labels, yticklabels=labels)\n",
    "    plt.xlabel('Predicted Class')\n",
    "    plt.ylabel('Original Class')\n",
    "    plt.title(\"Precision matrix\")\n",
    "    \n",
    "    plt.subplot(1, 3, 3)\n",
    "    # representing B in heatmap format\n",
    "    sns.heatmap(A, annot=True, cmap=cmap, fmt=\".3f\", xticklabels=labels, yticklabels=labels)\n",
    "    plt.xlabel('Predicted Class')\n",
    "    plt.ylabel('Original Class')\n",
    "    plt.title(\"Recall matrix\")\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train confusion_matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test confusion_matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Train confusion_matrix')\n",
    "plot_confusion_matrix(y_train,y_train_pred)\n",
    "print('Test confusion_matrix')\n",
    "plot_confusion_matrix(y_test,y_test_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "fpr,tpr,ths = roc_curve(y_test,y_test_pred)\n",
    "auc_sc = auc(fpr, tpr)\n",
    "plt.plot(fpr, tpr, color='navy',label='ROC curve (area = %0.2f)' % auc_sc)\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver operating characteristic with test data')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features = df_final_train.columns\n",
    "importances = clf.feature_importances_\n",
    "indices = (np.argsort(importances))[-25:]\n",
    "plt.figure(figsize=(10,12))\n",
    "plt.title('Feature Importances')\n",
    "plt.barh(range(len(indices)), importances[indices], color='r', align='center')\n",
    "plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
    "plt.xlabel('Relative Importance')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building XGBOOST model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean test scores [0.98055867 0.98054863 0.98104978 0.98098559 0.98127576]\n",
      "mean train scores [0.99999503 1.         0.99469271 0.99720075 0.99748026]\n"
     ]
    }
   ],
   "source": [
    "import xgboost as xgb\n",
    "clf = xgb.XGBClassifier()\n",
    "param_dist = {\"n_estimators\":sp_randint(105,125),\n",
    "              \"max_depth\": sp_randint(10,15)\n",
    "              }\n",
    "model = RandomizedSearchCV(clf, param_distributions=param_dist,\n",
    "                                   n_iter=5,cv=3,scoring='f1',random_state=25)\n",
    "\n",
    "\n",
    "model.fit(df_final_train,y_train)\n",
    "print('mean test scores',model.cv_results_['mean_test_score'])\n",
    "print('mean train scores',model.cv_results_['mean_train_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
      "       colsample_bynode=1, colsample_bytree=1, gamma=0, learning_rate=0.1,\n",
      "       max_delta_step=0, max_depth=11, min_child_weight=1, missing=None,\n",
      "       n_estimators=112, n_jobs=1, nthread=None,\n",
      "       objective='binary:logistic', random_state=0, reg_alpha=0,\n",
      "       reg_lambda=1, scale_pos_weight=1, seed=None, silent=None,\n",
      "       subsample=1, verbosity=1)\n"
     ]
    }
   ],
   "source": [
    "print(model.best_estimator_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf=xgb.XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
    "       colsample_bynode=1, colsample_bytree=1, gamma=0, learning_rate=0.1,\n",
    "       max_delta_step=0, max_depth=11, min_child_weight=1, missing=None,\n",
    "       n_estimators=112, n_jobs=1, nthread=None,\n",
    "       objective='binary:logistic', random_state=0, reg_alpha=0,\n",
    "       reg_lambda=1, scale_pos_weight=1, seed=None, silent=None,\n",
    "       subsample=1, verbosity=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf.fit(df_final_train,y_train)\n",
    "y_train_pred = clf.predict(df_final_train)\n",
    "y_test_pred = clf.predict(df_final_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train f1 score 0.9958231570552759\n",
      "Test f1 score 0.9269530001065757\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "print('Train f1 score',f1_score(y_train,y_train_pred))\n",
    "print('Test f1 score',f1_score(y_test,y_test_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train confusion_matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test confusion_matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Train confusion_matrix')\n",
    "plot_confusion_matrix(y_train,y_train_pred)\n",
    "print('Test confusion_matrix')\n",
    "plot_confusion_matrix(y_test,y_test_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "fpr,tpr,ths = roc_curve(y_test,y_test_pred)\n",
    "auc_sc = auc(fpr, tpr)\n",
    "plt.plot(fpr, tpr, color='navy',label='ROC curve (area = %0.2f)' % auc_sc)\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver operating characteristic with test data')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features = df_final_train.columns\n",
    "importances = clf.feature_importances_\n",
    "indices = (np.argsort(importances))[-25:]\n",
    "plt.figure(figsize=(10,12))\n",
    "plt.title('Feature Importances')\n",
    "plt.barh(range(len(indices)), importances[indices], color='r', align='center')\n",
    "plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
    "plt.xlabel('Relative Importance')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------------+--------------+-----------+----------------+---------------+\n",
      "|     Model     | n_estimators | max_depth | Train f1-Score | Test f1-Score |\n",
      "+---------------+--------------+-----------+----------------+---------------+\n",
      "| Random Forest |     121      |     14    |     0.9648     |     0.9149    |\n",
      "|    XGBOOST    |     112      |     11    |     0.9958     |     0.9269    |\n",
      "+---------------+--------------+-----------+----------------+---------------+\n"
     ]
    }
   ],
   "source": [
    "from prettytable import PrettyTable \n",
    "ptable = PrettyTable()\n",
    "ptable.field_names = [\"Model\", \"n_estimators\", \"max_depth\", \"Train f1-Score\",\"Test f1-Score\"]\n",
    "ptable.add_row(['Random Forest','121','14','0.9648','0.9149'])\n",
    "ptable.add_row(['XGBOOST','112','11','0.9958','0.9269'])\n",
    "print(ptable)"
   ]
  }
 ],
 "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": "3.5.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}