IBM_Personality Graphs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:19.329537Z",
     "start_time": "2018-10-26T20:19:15.439687Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import json\n",
    "%pylab inline\n",
    "import seaborn as sns\n",
    "import re\n",
    "with open('personality_output.json','r') as fp:\n",
    "    personality_output = json.load(fp)\n",
    "binny = pd.read_csv('selected_personality.txt',delimiter = '\\n',dtype = str,header = None)[0].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:28.701982Z",
     "start_time": "2018-10-26T20:19:19.333346Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABCwAAAC2CAYAAAASoBymAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X9wVPW9//HXJiEaCiSBC5tUQ+oUVBx+RL8txYsml40hJYEqEjoqV7lfm2ak0kuvWJSvNlospIJYaB3DzeQmeIlgBa6NZL34A8TIhSIz1omYjlycGxt/JHIFMciPwGa/fzhZsyRwDtn97NndPB8zzORkzzn73ldjmvM+5/P5uPx+v18AAAAAAABRJMHpAgAAAAAAAM5FwwIAAAAAAEQdGhYAAAAAACDq0LAAAAAAAABRh4YFAAAAAACIOjQsAAAAAABA1KFhAQAAHLF06VJdf/31mjlzZp+v+/1+/eY3v1FBQYFmzZql9957L8IVxjbyNY+MzSJf88jYLPINHQ0LAADgiFtvvVXV1dXnfb2xsVEtLS165ZVX9Nhjj+nRRx+NXHFxgHzNI2OzyNc8MjaLfENHwwIAADji+9//vlJTU8/7+o4dO3TLLbfI5XIpJydHX375pT777LMIVhjbyNc8MjaLfM0jY7PIN3RJThdwrsOHO5wuISxGjhzqdAnnFQ8Zk695ZGwW+ZpFvuZFIuP29nZlZGQEtjMyMtTe3q5Ro0Zd8Di/3y+Xy2W6vJhHvuaRsVnka15/MiZf+/gZthZ1DQsA8aOxsVHLly9XV1eX5s6dq7KysqDXP/nkEz3wwAPq6OiQz+fT/fffr7y8PIeqBRBt/H5/r+/Z+QPN5XLFfGMoEg2hgZyvRMamka9ZkWrM9yfjeMhX4mfYNLv50rAAYITP59OyZctUW1srt9utkpISeTwejRkzJrBPZWWlZsyYoTvuuEOHDh1SWVmZdu7c6WDVAKJJRkaG2traAtttbW2Wd51gH/maR8Zmka95ZGwW+VpjDgsARjQ1NSk7O1tZWVlKTk5WcXGxduzYEbSPy+XS8ePHJUkdHR38ggYQxOPx6E9/+pP8fr/eeecdDR06lN8TYUS+5pGxWeRrHhmbRb7WeMICgBHnjslzu91qamoK2mfhwoX6yU9+orq6Op08eVK1tbW2zp2ePlhJSYlhrRdA5N1333166623dPToUeXm5urnP/+5zp49K0m6/fbblZeXpzfeeEMFBQVKSUnRihUrHK44tpCveWRsFvmaR8ZmkW/oXP6+Bs44KNbH4nRjwjezyNe8UDP+z//8T+3evVvLly+XJP3pT3/Su+++q1/96leBfWpra+X3+3X33XfrL3/5ix566CE1NDQoIeHCD3/FQ8b8DJtFvuZFc8ZS7OdMvuaRsVnkaxb5mkfGZtnNd8ANCampqdJtt92impoqp0uJW2Qc3SL1v8+5Y/L6mvF4y5YtmjFjhiTp2muv1enTp3X06FGjdUUC/w2YRb7mkTEAAIgGA6phcerUSb366n9Kkl59dbtOnTrpcEXxh4yjWyT/95kwYYJaWlrU2tqqzs5Oeb1eeTyeoH0yMzO1d+9eSdIHH3yg06dPa/jw4cZqigT+GzCLfM0jYwAAEC0G1BwWZ86cCSwd4/d36cyZM7r00hSHq4ovZBzdIvm/T1JSksrLy1VaWiqfz6c5c+Zo7NixWrt2rcaPH6/8/Hw9+OCDevjhh7V+/Xq5XC799re/jfk1pflvwCzyNY+MAQBAtBhQDQvgXI2NjVq+fLm6uro0d+5clZWVBb3+H//xH1q5cqXcbrck6R//8R81d+5cJ0qNSXl5ecrLywv63qJFiwJfjxkzRs8991ykywIAAAAQA2KuYVFTU6VXXnlJ06cX6e67y6wPAM7D5/Np2bJlqq2tldvtVklJiTwej8aMGRO0X1FRkcrLyx2qEtFs/+J/7vW9kz5f0PZffrVUKYnfrGjy/dW/N14XAAAAEA9iqmFx7rjaO+64k8dUHfTLhod7fc93+mzQ9iOvrFDiJd/8mK2a+RvL80aqKdXU1KTs7GxlZWVJkoqLi7Vjx45eDQsAAAAAQOTFVMPiYsfVnntBbXUxLdm7oI6kgTZkIZJNqfb2dmVkZAS23W63mpqaeu33yiuvaP/+/briiiu0dOlSZWZmXvC86emDlZSUeMF9nJKc3BW0PWLEEKWmRveSTQAAAAAGpphqWAw0A3HIQiQne+t+n57OnfBx2rRpmjlzppKTk7Vp0yY98MAD+vd///cLnvfo0RNhrTOcOjqOB21//vlxdXb2vVhQtK89Ha0Se/wMuc7ZBgAAAGAfDYsoxpAFszIyMtTW1hbYbm9v16hRo4L2SU9PD3z94x//WE888UTE6guHc+dYsJpfQWKOhVAlJyQo51tD9M5XxzXpW0OUnDCgVo8GAAAAwoaGRRSLxSELroQed5Nd52zL+q59JIcsTJgwQS0tLWptbZXb7ZbX69Xq1auD9vnss88CTYydO3fqu9/9rpFaED7RMDFvftpw5acNd+S9443TTTerYXndtm/frkWLFmnLli2aMGFC2N7fNCaOBQAA0cxWw8LqD7ba2lpt3rxZiYmJGj58uFasWKHLLrtMkjRu3DhdeeWVkqTMzEytW7cuzB8hfsXikIWEQYkacuVwHT94REPGDlfCoOALicOHOy54vN0hC+EYrpCUlKTy8nKVlpbK5/Npzpw5Gjt2rNauXavx48crPz9fGzZs0M6dO5WYmKjU1FRVVFSE/L4wh4l5EU52h+UdP35cGzZs0KRJkxyqFAAAID5ZNizs/ME2btw4bd26VSkpKdq4caNWrVqlNWvWSJIuvfRS1dfXm/sEF8Hq7n+0cWLIQjjuTqdP/rbSJ3/bcr/+3NmTwnt3Ly8vT3l5eUHfW7RoUeDrxYsXa/HixWF7P5gVyTlQEP/sDstbu3atSktLVVNT40SZAAAAccuyYWHnD7YpU6YEvs7JydGLL75ooNTQWd39jzaRHrLA3WkgOoaUIDrYGZbX3NystrY2TZs2zXbDIppXErKDCXkBAECkWDYs7M6j0G3Lli3Kzc0NbJ8+fVq33nqrkpKSVFZWpptuuumC73ehP+TCMb+B1d3/aPpDLNJDFrg7jYGOph16shqW19XVpYqKiov+vRvNKwnZ0T20L5r+/xIAAMQny4aFnXkUutXX1+vAgQOqq6sLfO/111+X2+1Wa2ur5s+fryuvvFKjR48+7/td6A+5i1mSsb+s5liwK1x/yDFkAeHEkpsXRtMOPVkNy/vqq6908OBB3XXXXZKkw4cPa8GCBaqsrIypiTcBAACilWXDws48CpK0Z88erVu3TnV1dUpOTg583+12S5KysrI0efJkNTc3X7BhAcCceFpyc9Gq3kPPus6eCtr+f09tV0LSpYHttb/8UdDrv2x4OGjbd/ps0PYjr6xQ4iXBvyZXzfxNv+pF7LEaljd06FDt27cvsH3nnXdqyZIlNCsAAADCxPJqpecfbJ2dnfJ6vfJ4PEH7NDc3q7y8XJWVlRoxYkTg+8eOHVNnZ6ck6ciRI3r77bd7TVYWi2pqqnTbbbeopqbK6VLiDk8AmJefNlyLLxvNspuAhZ7D8oqKijRjxozAsLwdO3Y4XZ4x/B4GAADRwvIJCzvzKKxcuVInTpwIDFXoXr70gw8+0COPPCKXyyW/36+f/vSnMd+wYIy7WfH0BABiT6ytJDQQRfpi2mpYXk8bNmwwWkuk8HsYAABEC8uGhWT9B9v69ev7PO66667Ttm3b+lVYOB73NoEx7ublpw3n7j8cEWsrCQ1EXExHBr+HAQBANLDVsACAmODq2WBwnbNtj9VKQnAeF9MAAAADA7emAMSNhMRBShk5TpKUMvJqJSQOcrgiAFYaGxtVWFiogoICVVX1nhvqk08+0Z133qlbbrlFs2bN0htvvOFAlbGLfM0iX/PI2CzyNY+MQ8MTFgNUf4bcSJEZdgOEYtjo6zVs9PVOl4ELqKmp0iuvvKTp04t0991lTpcDB/l8Pi1btky1tbVyu90qKSmRx+MJmu+qsrJSM2bM0B133KFDhw6prKxMO3fudLDq2EG+ZpGveWRsFvmaR8ah4wkLAEDEnDtx8alTJx2uCE5qampSdna2srKylJycrOLi4l4rsLhcLh0/flyS1NHR0efS6ugb+ZpFvuaRsVnkax4Zh44nLAAAEcPExeipvb1dGRkZgW23262mpqagfRYuXKif/OQnqqur08mTJ1VbW2vr3CNHDg1rrbGIfM0yma9ExhI/w6aRr3lkHDoaFhewf/E/9/reSZ8vaPsvv1qqlMRvJvb7/urfG68LALo1NjZq+fLl6urq0ty5c1VWFjzEYtOmTdq4caMSEhI0ePBgPfbYYzG/vDTiR3fzqifXOUvVer1ezZ49W3fffbf+8pe/aMmSJWpoaFCCxQoxhw93hLXWSAvHH6Lke2GhZmwyXyn2M+Zn2CzyNY+MzbKbL0NCACBGdY+LrK6ultfrVUNDgw4dOhS0z6xZs7Rt2zbV19ertLRUFRUVDlUL9JaRkaG2trbAdnt7e69HYbds2aIZM2ZIkq699lqdPn1aR48ejWidsYp8zSJf88jYLPI1j4xDR8MCiDE1NVW67bZbVFPTe5ZhDCx2xkUOGTIk8PXJkyd7dfVN+mXDw73+PfLKiqB9HnllRdDrGFgmTJiglpYWtba2qrOzU16vVx6PJ2ifzMxM7d27V5L0wQcf6PTp0xo+nGVt7SBfs8jXPDI2i3zNI+PQMSQEiCHnTlh4xx13Mv5/ALMzLlKSnn32WdXW1urMmTN65plnLM+bnj5YSUmJlvuZEI7HLwfKmM54kJSUpPLycpWWlsrn82nOnDkaO3as1q5dq/Hjxys/P18PPvigHn74Ya1fv14ul0u//e1vI9p4i2Xkaxb5mkfGZpGveWQcOhoWQAxhwkL0ZGdcpCTNmzdP8+bN07Zt21RZWanHH3/8guc9evRE2Gq8WOEYj3n4cAdNixiSl5envLy8oO8tWrQo8PWYMWP03HPPRbqsuEG+ZpGveWRsFvmaR8ahYUgIvuHqeUfVdc42gGhjZ1xkT8XFxXrttdciURoAAAAQsthqWETBBXVij7uXrnO2Y11C4iCljBwnSUoZebUSEgc5XJF5jY2NKiwsVEFBgaqqzj8nxPbt23XVVVfp3XffjWB1sc9Ovi+99JKKiopUXFysxYsXR7jC2GZnXGRLS0vg6127dik7OzvCVQIAAAD9E1NDQrovqE8e/qtjF9TJCQnK+dYQvfPVcU361hAl21iWKpYMG329ho2+3ukyIqJ7hYXa2lq53W6VlJTI4/H0WvLx+PHj2rBhgyZNmuRQpbHJTr4tLS2qqqrSpk2blJqaqs8//9zBimOPnXGRdXV12rt3r5KSkjRs2DDL4SCmuRJ6NHld52wDAAAAPcRUw0KKjgvq/LThyk+LzMytjY2NWr58ubq6ujR37lyVlZX1ud/27du1aNEibdmyRRMmTIhIbbGu5woLkgIrLJzbsFi7dq1KS0tVU1MT0fr6WjHBd/ps0PYjr6xQ4iXf/Ge8auZvjNdll518n3/+ec2bN0+pqamSpBEjRjhSayyzGhf58MPRtfJGwqBEDblyuI4fPKIhY4crYRBDz8KtpqZKr7zykqZPL9Ldd/f9/xkAAACxIOYaFgMJTwCYZWeFhebmZrW1tWnatGm2GxassPA1O/l2D1e47bbb1NXVpYULFyo3N9fy3Onpg/tVUzTkw2SQUvrkbyt98redLiMusZIQAACIJzQsoli0PwEQ66xWWOjq6lJFRYUqKiou6rzxsMKCFPqFtZ0VLHw+nz788ENt2LBBbW1tmjdvnhoaGjRs2LALnru/GYczn1CPj9XGBXfvoxsrCQEAgHhiawIGq4nzamtrVVRUpFmzZmn+/Pn6+OOPA6+98MILmj59uqZPn64XXnghfJUPAH3doW5vbw/ap+cTAHb19+609PVFVvc/J47vPkc4WK2w8NVXX+ngwYO666675PF49M4772jBggVMvGmTnRUs3G638vPzNWjQIGVlZemKK64ImiQS0eXcu/enTp10uCIAAADEM8uGRfewhOrqanm9XjU0NOjQoUNB+4wbN05bt27Vtm3bVFhYqFWrVkmSvvjiCz311FN6/vnntXnzZj311FM6duyYmU8Sh+w+AfDAAw9c1HlDeQLg8OGOwD8nju8+RzhYrbAwdOhQ7du3Tzt37tTOnTuVk5OjyspKR+cIiaUJC+2sYHHTTTdp3759kqQjR46opaUl8EQRok9fd+8BAAAAUywbFj2HJSQnJweGJfQ0ZcoUpaR8/chpTk5O4K7q7t27NXXqVKWlpSk1NVVTp07Vm2++aeBjxCeeADCr5woLRUVFmjFjRmCFhXN/xqNF94SFkqJ+wkI7+d54441KS0tTUVGR5s+fryVLlig9Pd3hygEAAABEA8s5LOxMnNfTli1bApPm2RnScK5QJiyMtwn1et6hdrvd8nq9Wr16deD17icAut15551asmQJq4RcBKsVFnrasGFDyO8XjvH/sTRhoVW+LpdLS5cu1dKlSyNdGhDz+rOSkBRdqwkBAABciOUTFnYmzutWX1+vAwcOqLS09KKP7Xb06ImwDDdwashCqMMdeorFJwBwfoz/B2BHTU2VbrvtFtXU9J4zCgAAYCCxfMLCzsR5krRnzx6tW7dOdXV1Sk5ODhz71ltvBR07efLkcNQ9YET6CQCYw+z9QOxpbGzU8uXL1dXVpblz56qsLPjJqE2bNmnjxo1KSEjQ4MGD9dhjj/VayelisCwpAADANyyfsLAzcV5zc7PKy8tVWVmpESNGBL5/ww03aPfu3Tp27JiOHTum3bt364Ybbgj/pwAAIMzsTDo9a9Ysbdu2TfX19SotLb3oZZDPxcSmAAAA37B8wqLnsASfz6c5c+YEhiWMHz9e+fn5WrlypU6cOBG485+Zmal169YpLS1NP/vZz1RSUiJJuvfee5WWlmb2EwEAEAY9J52WFJh0uucTFEOGDAl8ffLkScthjwAAALDPsmEhWQ9LWL9+/XmPLSkpCTQsAACIFXYnnX722WdVW1urM2fO6Jlnnolkib3E0tLHAAAAVmw1LAAAGGjsThw9b948zZs3T9u2bVNlZaUef/zxC573QqthJSd3BW2PGDFEqan2V5/qXvr4+MEj5136OJpWwwIAALgQGhYAAPTB7qTT3YqLi/Xoo49anvfo0RPnfa2j43jQ9uefH1dnp+V0U0Gslj4OdSWr7uNpXAAAANNoWAAA+rRo1YtB211nTwVt/7+ntish6dKg76395Y+M1xUpPSeddrvd8nq9Wr16ddA+LS0t+s53viNJ2rVrl7Kzsx2oFAAAID7RsAAAoA92Jp2uq6vT3r17lZSUpGHDhlkOBwEAAIB9NCwAQwb63WkgHlhNOv3www9HuiQAAIAB4+IGxgIAAIRRY2OjCgsLVVBQoKqqqj73eemll1RUVKTi4mItXrw4whXGNvI1i3zNI2OzyNcs8g0dT1gAAABH+Hw+LVu2TLW1tXK73SopKZHH49GYMWMC+7S0tKiqqkqbNm1SamqqPv/8cwcrji3kaxb5mkfGZpGvWeQbHjxhAQAAHNHU1KTs7GxlZWUpOTlZxcXF2rFjR9A+zz//vObNm6fU1FRJ0ogRI5woNSaRr1nkax4Zm0W+ZpFvePCEBQAADjh3nhvJeq6beJvnpr29XRkZGYFtt9utpqamoH1aWlokSbfddpu6urq0cOFC5ebmWp6bZVfJ1zST+UpkLPEzbBr5msXviPCgYQEAABzh9/t7fc/lcgVt+3w+ffjhh9qwYYPa2to0b948NTQ0aNiwYRc89+HDHWGtNdLC8Yco+V5YqBmbzFeK/Yz5GTaLfM3jd4RZdvNlSAgAAHBERkaG2traAtvt7e0aNWpU0D5ut1v5+fkaNGiQsrKydMUVVwTuSOHCyNcs8jWPjM0iX7PINzxoWAAAAEdMmDBBLS0tam1tVWdnp7xerzweT9A+N910k/bt2ydJOnLkiFpaWpSVleVEuTGHfM0iX/PI2CzyNYt8w4MhIRjQGhsbtXz5cnV1dWnu3LkqKysLen3Tpk3auHGjEhISNHjwYD322GNBM/sCAPovKSlJ5eXlKi0tlc/n05w5czR27FitXbtW48ePV35+vm688Ub913/9l4qKipSYmKglS5YoPT3d6dJjAvmaRb7mkbFZ5GsW+YYHDQsMWHaWGpo1a5Zuv/12SdKOHTtUUVGhf/u3f3OqZMBZrsSeG+dsA/2Tl5envLy8oO8tWrQo8LXL5dLSpUu1dOnSSJcWF8jXLPI1j4zNIl+zyDd0DAmJco2NjSosLFRBQYGqqqp6vb5p0ybNmjVLN998s26//XYdOnTIgSpjk52lhoYMGRL4+uTJk70myrkoXOwhxiUkDlLKyHGSpJSRVyshcZDDFQEAACCe8YRFFOMJALPsLDUkSc8++6xqa2t15swZPfPMM5bnTU8frKSk3s2I7ou9k4f/et6LvVBnIw7HjNEDZYkk9M+w0ddr2OjrnS4DAAAAA4CtJyys7vLv379fs2fP1jXXXKPt27cHvTZu3DjdfPPNuvnmm3XPPfeEp+oBIuJPAAwwdpYakqR58+bptdde0/3336/KykrL8x49euK8ywwNG3293P/n7vNe8B0+3BH41x89jw/1HAAAAADgJMsnLOzc5c/MzFRFRYVqamp6HX/ppZeqvr4+vFUPECafAOiveHoCwM5SQz0VFxfr0UcfDct7A+FiNXFsbW2tNm/erMTERA0fPlwrVqzQZZdd5lC1AAAAgH2WT1jYuct/+eWX6+qrr1ZCAlNihJPJJwD6K5xPAIRSQzjYWWqo5zrIu3btUnZ2dljeGwiH7oZydXW1vF6vGhoaes1jM27cOG3dulXbtm1TYWGhVq1a5VC1sIW5bgAAAAIsn7Cwe5f/fE6fPq1bb71VSUlJKisr00033XTB/c83/t+OaLh7H87x/zwBYJadpYbq6uq0d+9eJSUladiwYXr88cedLhsI6NlQlhRoKPd8Am7KlCmBr3NycvTiiy9GvE7YZ2euGwAAgIHCsmFh9y7/+bz++utyu91qbW3V/PnzdeWVV2r06NHn3T/Uu/+hCvUc3ceHo3HR8wkAt9str9er1atXB+3T0tKi73znO5J4AqA/rJYaevjhhyNdEmDbxTaUt2zZotzcXMvzOtk4jobGs9OY2BQAAOBrlg2Li73Lfy632y1JysrK0uTJk9Xc3HzBhgW+wRMAiHVW8yt02759uxYtWqQtW7ZowoQJEa4ydl1MQ7m+vl4HDhxQXV2d5XmdbByHq/Ec600LAAAA2GhY2LnLfz7Hjh1TSkqKkpOTdeTIEb399tsqLS0NueiBhCcAEKvsTNgrScePH9eGDRs0adIkhyqNXXYbynv27NG6detUV1en5OTkSJYIAAAA9JvlLJk97/IXFRVpxowZgbv83ZNvNjU1KTc3V9u3b9cjjzyi4uJiSdIHH3ygOXPm6Ec/+pHmz5+vn/70p70uVgDEJzsT9krS2rVrVVpaqksuucSBKmObnYljm5ubVV5ersrKSo0YMcKhSgEAAICLZ/mEhWR9l3/ixIlqbGzsddx1112nbdu2hVgigFhkZ36F5uZmtbW1adq0aX0ui3w+/V2aNxrmRwjnUAU7w8ZWrlypEydOBH5nZ2Zmat26dWGrAQAAADDFVsMCAC6W1fwKXV1dqqioUEVFxUWfu79zLMTbxLySdUN5/fr1YXkfAAAAINIsh4QAQH9Yza/w1Vdf6eDBg7rrrrvk8Xj0zjvvaMGCBXr33XedKBcAAABAlOEJCwBGWE3YO3ToUO3bty+wfeedd2rJkiWsEgIAAABAEk9YADDEzoS9QLRrbGxUYWGhCgoKVFVV1ev12tpaFRUVadasWZo/f74+/vhjB6oEAACITzxhAcAYq/kVetqwYUMkSgJss7M077hx47R161alpKRo48aNWrVqldasWeNg1QAAAPGDJywAAOiDnaV5p0yZopSUFElSTk5O0LwtAAAACA1PWAAA0Ac7S/P2tGXLFuXm5lqeNz19sJKSEvtVU7wtzQsAAHAhNCwAAOiD1dK8PdXX1+vAgQOqq6uzPG9/l+WV4nNpXgAAgPNhSAgAAH2wWpq32549e7Ru3TpVVlYqOTk5kiXGBauJTbtt375dV111FUsfXyTyNYt8zSNjs8jXPDIODQ0LAAD60HNp3s7OTnm9Xnk8nqB9mpubVV5ersrKSo0YMcKhSmNX98Sm1dXV8nq9amho0KFDh3rtd/z4cW3YsEGTJk1yoMrYRb5mka95ZGwW+ZpHxqGjYQEAQB/sLM27cuVKnThxQosWLdLNN9+se+65x+GqY4udiU0lae3atSotLdUll1ziQJWxi3zNIl/zyNgs8jWPjEPHHBYAAJyH1dK869evj3BF8cXOxKbNzc1qa2vTtGnTVFNTY/vczLFBvqaZzFciY4mfYdPI1zwyDh0NCwAA4AiriU27urpUUVGhioqKiz53OCYodVI4/hAl3wsLNWOT+UqxnzE/w2aRr3lkbJbdfBkSggHNahKc2tpaFRUVadasWZo/f74+/vhjB6oEgPhkNbHpV199pYMHD+quu+6Sx+PRO++8owULFjAhmU3kaxb5mkfGZpGveWQcOhoWUY4LanPsTIIzbtw4bd26Vdu2bVNhYaFWrVrlULUAEH+sJjYdOnSo9u3bp507d2rnzp3KyclRZWWlJkyY4GDVsYN8zSJf88jYLPI1j4xDR8MiinFBbZadSXCmTJmilJQUSVJOTk5QhxQAEBo7E5ui/8jXLPI1j4zNIl/zyDh0tuawaGxs1PLly9XV1aW5c+eqrKws6PX9+/drxYoVev/99/Xkk0/qhz/8YeC1F154QZWVlZKkBQsWaPbs2WEsP771vKCWFLigHjNmTGCfKVOmBL7OycnRiy++GPE6Y5WdSXB62rJli3Jzcy3Pm54+WElJif2qKdSxcuEYazdQJvABEB2sJjbtacOGDZEoKa6Qr1nkax4Zm0W+5pFxaCwbFt13+Wtra+V2u1VSUiKPxxN00ZyZmakK/3syAAARLklEQVSKiopes5p+8cUXeuqpp7R161a5XC7deuut8ng8Sk1NDf8niUMmL6j7K54uqK0mwempvr5eBw4cUF1dneV5jx490e+aQp08JxyT73Sfg8YFAAAAACdZNizs3OW//PLLJUkJCcEjTHbv3q2pU6cqLS1NkjR16lS9+eabmjlzZtg+QDzjgvr85wjHxbTVJDjd9uzZo3Xr1qmurk7Jyckhvy8AAAAAwJplw+Ji7/JbHdve3n7BY5x8nD4c5wjnXWkuqM3qOQmO2+2W1+vV6tWrg/Zpbm5WeXm5qqurNWLECIcqBQAAAICBx7JhcTF3+cNxrJN3/8NxjnA+Ts8FtVk9J8Hx+XyaM2dOYBKc8ePHKz8/XytXrtSJEycC48wyMzO1bt06hysHAAAAgPhn2bCwe5f/fMe+9dZbQcdOnjy5H2UOTFxQm2c1Cc769esjXBEAAAAAQLLRsLBzl/98brjhBj355JM6duyYpK/ntLjvvvtCq3iA4YIaAAAAADAQWTYs7Nzlb2pq0sKFC/Xll1/q9ddf1x/+8Ad5vV6lpaXpZz/7mUpKSiRJ9957b2ACTgAAAAAAgPOxbFhI1nf5J06cqMbGxj6PLSkpCTQsAAAAAAAA7Eiw3gUAAAAAACCyaFgAAAAAAICoQ8MCAAAAAABEHRoWAAAAAAAg6tCwAIAY1tjYqMLCQhUUFKiqqqrX6/v379fs2bN1zTXXaPv27Q5UCAAAAPQPDQsAiFE+n0/Lli1TdXW1vF6vGhoadOjQoaB9MjMzVVFRoZkzZzpUJQAAANA/NCwAGGN197+2tlZFRUWaNWuW5s+fr48//tiBKmNXU1OTsrOzlZWVpeTkZBUXF2vHjh1B+1x++eW6+uqrlZDAr3sAAADEliSnCwAQn7rv/tfW1srtdqukpEQej0djxowJ7DNu3Dht3bpVKSkp2rhxo1atWqU1a9Y4WHVsaW9vV0ZGRmDb7Xarqakp5POmpw9WUlJiv44dOXJoSO8d6vHhOgcAAACcR8MCgBE97/5LCtz979mwmDJlSuDrnJwcvfjiixGvM5b5/f5e33O5XCGf9+jRE/0+9vDhjpDeO9Tju89B0wIAACD28YwwACP6uvvf3t5+3v23bNmi3NxcW+dOTx/cr5pGjhwa9C/Uc4RyfDhkZGSora0tsN3e3q5Ro0aF5dz4GpOaAgAAOIeGBQAjLubuf319vQ4cOKDS0lJb5+7vEwCHD3cE/Qv1HKEcHw4TJkxQS0uLWltb1dnZKa/XK4/HE5Zzg0lNAQAAnMaQEABG2L37v2fPHq1bt051dXVKTk6OZIkxLykpSeXl5SotLZXP59OcOXM0duxYrV27VuPHj1d+fr6ampq0cOFCffnll3r99df1hz/8QV6v1+nSY4KdYU2XX365JDGpKQAAgAE0LAAY0fPuv9vtltfr1erVq4P2aW5uVnl5uaqrqzVixAiHKo1teXl5ysvLC/reokWLAl9PnDhRjY2NkS4rLsTjpKbhOEe45wdpbGzU8uXL1dXVpblz56qsrCzo9draWm3evFmJiYkaPny4VqxYocsuuyysNcQz8jWLfM0jY7PI1yzyDR23hAAY0fPuf1FRkWbMmBG4+9+99ObKlSt14sQJLVq0SDfffLPuueceh6sGvmFyUlOnhiSF4xzhHNZkZ9hN92pC27ZtU2FhoVatWhWW9x4IyNcs8jWPjM0iX7PINzxoWGBAY0I9s/Ly8vTyyy/rtdde04IFCyR9ffc/Pz9fkrR+/Xrt2bNH9fX1qq+v17p165wsFwjCpKbm9Rx2k5ycHBh209OUKVOUkpIi6evVhHr+b4ILI1+zyNc8MjaLfM0i3/BgSEiUs3qMaP/+/VqxYoXef/99Pfnkk/rhD3/oUKWxp7vrWVtbK7fbrZKSEnk8nqDx6d0T6tXU1DhYKQAn2BnWhNBc7LCbi1lNiKVtydc0k/lKZCzxM2wa+ZrF74jwsNWwsLpo7uzs1JIlS/Tee+8pLS1Nv/vd73T55Zfro48+UlFRka644gpJ0qRJk7Rs2bLwf4o4xQW1WUyoB+BCmNTUvP6sJlRXV2fr3OEatuKUcPwhSr4XFmrGJvOVYj9jfobNIl/z+B1hlt18LRsWdi6aN2/erGHDhunVV1+V1+vVE088oTVr1kiSRo8erfr6+n5+jIGNC2qzTE2oByB+MKmpWawmZBb5mkW+5pGxWeRrFvmGh+VVrp2xNzt37tTs2bMlSYWFhdq7d2+fHSVcnL4uqNvb20M+b3r64H4fO3Lk0MA/J47vPkc4mJpQLz19sGP59Dw+1HMAgGk9h910dnbK6/XK4/EE7dO9mlBlZSWrCV0k8jWLfM0jY7PI1yzyDQ/LJyzs3IVub29XZmbm1ydMStLQoUN19OhRSdJHH32kW265RUOGDNEvfvELfe9737vg+7Hc2zdMzlDfX6E+ehSOR5cOH+4IS86mJtSLh3ylgTMuDoBz7Ay76bmakPT1UEgm6LWHfM0iX/PI2CzyNYt8w8OyYWHnovl8+4waNUqvv/660tPTdeDAAd17773yer0aMmTIed/PyYu9cJwjnBd7zFBvFhPqAYDzrIbdrF+/PsIVxRfyNYt8zSNjs8jXLPINneWQEDsXzRkZGfr0008lSWfPnlVHR4fS0tKUnJys9PR0SdL48eM1evRo/c///E84649rdh4jQv/17HoWFRVpxowZga5n97CnpqYm5ebmavv27XrkkUdUXFzscNUAAAAAMDBYPmFh5y60x+PRCy+8oGuvvVYvv/yypkyZIpfLpSNHjig1NVWJiYlqbW1VS0tLYAJJWGOGevOYUA8AAAAAopNlw8LORXNJSYl++ctfqqCgQKmpqfrd734nSdq/f79+//vfKzExUYmJifr1r3+ttLQ04x8qnnBBDQAAAAAYiCwbFpL1RfMll1yi3//+972OKywsVGFhYYglAgAAAACAgcZyDgsAAAAAAIBIo2EBAAAAAACiDg0LAAAAAAAQdWhYAAAAAACAqEPDAgAAAAAARB0aFgAAAAAAIOrQsAAAAAAAAFGHhgUAAAAAAIg6NCwAAAAAAEDUoWEBAAAAAACiDg0LAAAAAAAQdWhYAAAAAACAqEPDAgAAAAAARB0aFgAAAAAAIOrQsAAAAAAAAFGHhgUAAAAAAIg6thoWjY2NKiwsVEFBgaqqqnq93tnZqV/84hcqKCjQ3Llz9dFHHwVe+9d//VcVFBSosLBQb775ZvgqHyBCyR7WyNcs8jWPjM0iX/PI2CzyNYt8zSNjs8jXPDIOjWXDwufzadmyZaqurpbX61VDQ4MOHToUtM/mzZs1bNgwvfrqq/qnf/onPfHEE5KkQ4cOyev1yuv1qrq6Wr/+9a/l8/nMfJI4FEr2sEa+ZpGveWRsFvmaR8Zmka9Z5GseGZtFvuaRcegsGxZNTU3Kzs5WVlaWkpOTVVxcrB07dgTts3PnTs2ePVuSVFhYqL1798rv92vHjh0qLi5WcnKysrKylJ2draamJjOfJA6Fkj2ska9Z5GseGZtFvuaRsVnkaxb5mkfGZpGveWQcuiSrHdrb25WRkRHYdrvdvZoO7e3tyszM/PqESUkaOnSojh49qvb2dk2aNCno2Pb29gu+38iRQyVJG1fOs/8pzmP9/10b0vFF/14bcg2hCCX74cOHn/e8I0cOJV+ZzVcK/Wc41HwlZzM2la/Ez3A3fobNIl/zTP+eGOjI1yyT+UpkLPEzbBr5mkfGobN8wqKv7o7L5bK1j51jcX6hZA9r5GsW+ZpHxmaRr3lkbBb5mkW+5pGxWeRrHhmHzrJhkZGRoba2tsB2e3u7Ro0a1WufTz/9VJJ09uxZdXR0KC0tzdaxOL9Qsoc18jWLfM0jY7PI1zwyNot8zSJf88jYLPI1j4xDZ9mwmDBhglpaWtTa2qrOzk55vV55PJ6gfTwej1544QVJ0ssvv6wpU6bI5XLJ4/HI6/Wqs7NTra2tamlp0cSJE818kjgUSvawRr5mka95ZGwW+ZpHxmaRr1nkax4Zm0W+5pFxGPht2LVrl3/69On+/Px8/9NPP+33+/3+NWvW+F977TW/3+/3nzp1yv/zn//cf9NNN/nnzJnj/9vf/hY49umnn/bn5+f7p0+f7t+1a5edt0MPoWQPa+RrFvmaR8Zmka95ZGwW+ZpFvuaRsVnkax4Zh8bl9zMFKQAAAAAAiC6WQ0IAAAAAAAAijYYFAAAAAACIOjQsAABAXGhsbFRhYaEKCgpUVVXldDn9snTpUl1//fWaOXOm06X0KdYzJl/zojlj8jWPjM0aiPnSsAAAADHP5/Np2bJlqq6ultfrVUNDgw4dOuR0WRft1ltvVXV1tdNl9CkeMiZf86I1Y/I1j4zNGqj50rAAAAAxr6mpSdnZ2crKylJycrKKi4u1Y8cOp8u6aN///veVmprqdBl9ioeMyde8aM2YfM0jY7MGar40LAAAQMxrb29XRkZGYNvtdqu9vd3BiuIPGZtFvmaRr3lkbNZAzTemGxZXXXWVvvrqq6Dv/eAHP9BHH33kUEXxhXzNIl/zyNgs8jWPjO3ra5V2l8vlQCXxi4zNIl+zyNc8MjZroOYb0w0L03w+n9MlxDXyNYt8zSNjs8jXvHjKOCMjQ21tbYHt9vZ2jRo1ysGK4g8Zm0W+ZpGveWRs1kDNN8npAkzp6urSsmXL9Oc//1nJyckaPHiwnnvuOUnSG2+8ocrKSnV2dmrQoEFaunSpcnJytG/fPq1YsULf+9739O6772rBggX67LPPtH79eiUnJ6urq0tr1qzRd7/7XYc/nfPI1yzyNY+MzSJf88g42IQJE9TS0qLW1la53W55vV6tXr3a6bLiChmbRb5mka95ZGzWgM3XH8OuvPJK//Hjx4O+N3nyZH9ra6v/vffe80+fPt3v8/n8fr/f/8UXX/j9fr//ww8/9P/4xz/2d3R0+P1+v//gwYP+vLw8v9/v9//5z3/2X3311f633347cL7rrrvO/8knn/j9fr//9OnT/hMnTpj+WFGDfM0iX/PI2CzyNY+ML86uXbv806dP9+fn5/uffvppp8vpl3/5l3/xT5061X/NNdf4b7zxRv/zzz/vdElBYj1j8jUvmjMmX/PI2KyBmG9cPmHhcrmUlZUln8+nhx56SD/4wQ80bdo0SdKbb76pv/3tb5o3b15g/7Nnz+p///d/JUnZ2dm69tprA69NmTJFS5cuVX5+vv7hH/5BWVlZkf0wUYh8zSJf88jYLPI1j4z7lpeXp7y8PKfLCMmTTz7pdAkXFOsZk6950Zwx+ZpHxmYNxHxjeg6L4cOH64svvghsnz17VsePH9fw4cM1dOhQeb1ezZgxQ++//76Ki4t1+PBhSdKNN96o+vr6wL/du3fr7/7u7yRJgwcPDnqPp556Svfdd59Onjypu+66S2+88UbkPqDDyNcs8jWPjM0iX/PIGAAADGQx3bD4+7//e/3xj38MbP/xj3/UpEmTlJKSoiNHjujUqVPKzc3V/fffr6FDh6q1tVVTp07Vm2++qf/+7/8OHNfU1NTn+c+ePavW1lZNnDhRZWVlmjp1qv76178a/1zRgnzNIl/zyNgs8jWPjAEAwEAW00NCHnroIS1fvlyzZs1SQkKCMjMztXLlSknSp59+ql/96lc6e/asfD6fcnNzlZOTo4SEBK1atUoPPfSQTp06pTNnzui6667TxIkTe52/q6tLDz74oDo6OuRyuZSZmanFixdH+mM6hnzNIl/zyNgs8jWPjAEAwEDm8vv7WNAVAAAAAADAQTE9JAQAAAAAAMQnGhYAAAAAACDq0LAAAAAAAABRh4YFAAAAAACIOjQsAAAAAABA1KFhAQAAAAAAog4NCwAAAAAAEHX+PzNdN8K6uL1UAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x194.4 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set()\n",
    "# keys = pd.read_csv('selected_personality.txt',delimiter = '\\n',header = None).values.ravel().tolist()\n",
    "keys = ['_r_personality_parent','_r_personality_parent_Openness','_r_personality_parent_Conscientiousness'\n",
    "        ,'_r_personality_parent_Extraversion','_r_personality_parent_Agreeableness',\n",
    "        '_r_personality_parent_Neuroticism','_r_needs_parent','_r_values_parent']\n",
    "###############################\n",
    "hate_df = pd.DataFrame(np.c_[['hate' for i in range(len(list(personality_output.values())[0]['hate']))]]\n",
    "                            ,columns =['Users'] )\n",
    "counter_df = pd.DataFrame(np.c_[['counter' for i in range(len(list(personality_output.values())[0]['counter']))]]\n",
    "                            ,columns =['Users'] )\n",
    "Noncounter_df = pd.DataFrame(np.c_[['Noncounter' for i in range(len(list(personality_output.values())[0]['noncounter']))]]\n",
    "                            ,columns =['Users'] )\n",
    "c = 0\n",
    "for key,value in personality_output.items():\n",
    "    if key not in keys:\n",
    "        hate_df[key] = value['hate']\n",
    "        counter_df[key] = value['counter']\n",
    "        Noncounter_df[key] = value['noncounter']\n",
    "    else:\n",
    "        comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "        comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "        num_subplots = len(comb_df.columns[1:])\n",
    "        if num_subplots>=2:\n",
    "            f,axes = plt.subplots(1,12,figsize = (15,2.7))\n",
    "            for i,column in enumerate(comb_df.columns[1:]):\n",
    "                sns.barplot(x = 'Users',y =column,data = comb_df,ax = axes[i])\n",
    "    #                     axes[plot].set_yticklabels([])\n",
    "                axes[i].set_ylabel('')    \n",
    "#                 axes[i].set_xlabel(re.findall(r'[^_]+',column)[-1])\n",
    "    #                     axes[plot].set_size_inches(w=11,h=7)\n",
    "    #                     axes[plot].\n",
    "    #                     for tick in axes[plot].get_xticklabels():\n",
    "    #                         tick.set_rotation(45)\n",
    "                axes[i].set_xticklabels([])\n",
    "            plt.tight_layout()\n",
    "            plt.show()\n",
    "            hate_df = pd.DataFrame(np.c_[['hate' for i in range(len(list(personality_output.values())[0]['hate']))]]\n",
    "                                ,columns =['Users'] )\n",
    "            counter_df = pd.DataFrame(np.c_[['counter' for i in range(len(list(personality_output.values())[0]['counter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "            Noncounter_df = pd.DataFrame(np.c_[['Noncounter' for i in range(len(list(personality_output.values())[0]['noncounter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "            \n",
    "#################################### END PLOT ###############################################\n",
    "comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "num_subplots = len(comb_df.columns[1:])\n",
    "\n",
    "f,axes = plt.subplots(1,12,figsize = (15,2.7))\n",
    "#                 print(num_subplots)\n",
    "for i,column in enumerate(comb_df.columns[1:]):\n",
    "    sns.barplot(x = 'Users',y =column,data = comb_df,ax = axes[i])\n",
    "#                     axes[plot].set_yticklabels([])\n",
    "    axes[i].set_ylabel('')    \n",
    "#     axes[i].set_xlabel(re.findall(r'[^_]+',column)[-1])\n",
    "    axes[i].set_xticklabels([])\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:33.440587Z",
     "start_time": "2018-10-26T20:19:28.704501Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 194.4x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(palette = 'muted')\n",
    "# keys = pd.read_csv('selected_personality.txt',delimiter = '\\n',header = None).values.ravel().tolist()\n",
    "keys = ['_r_personality_parent','_r_personality_parent_Openness','_r_personality_parent_Conscientiousness'\n",
    "        ,'_r_personality_parent_Extraversion','_r_personality_parent_Agreeableness',\n",
    "        '_r_personality_parent_Neuroticism','_r_needs_parent','_r_values_parent']\n",
    "###############################\n",
    "hate_df = pd.DataFrame()\n",
    "counter_df = pd.DataFrame()\n",
    "Noncounter_df = pd.DataFrame()\n",
    "for key,value in personality_output.items():\n",
    "    if key not in keys:\n",
    "        temp_h = pd.DataFrame(np.c_[['hate' for i in range(len(list(personality_output.values())[0]['hate']))]]\n",
    "                                ,columns =['Users'])\n",
    "        temp_c = pd.DataFrame(np.c_[['counter' for i in range(len(list(personality_output.values())[0]['counter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "        temp_nc = pd.DataFrame(np.c_[['Noncounter' for i in range(len(list(personality_output.values())[0]['noncounter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "        temp_h['Value'],temp_c['Value'],temp_nc['Value'] = value['hate'],value['counter'],value['noncounter']\n",
    "        temp_h['Attributes'],temp_c['Attributes'],temp_nc['Attributes'] = re.findall(r'[^_]+',key)[-1],re.findall(r'[^_]+',key)[-1],re.findall(r'[^_]+',key)[-1]\n",
    "        hate_df = pd.concat([hate_df,temp_h],ignore_index = True)\n",
    "        counter_df = pd.concat([counter_df,temp_c],ignore_index = True)\n",
    "        Noncounter_df = pd.concat([Noncounter_df,temp_nc],ignore_index = True)\n",
    "    else:\n",
    "        try:\n",
    "            if len(set(hate_df['Attributes']))==12:\n",
    "                comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "                comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "#                 plt.figure(figsize = (3,3))\n",
    "                ax = sns.barplot(x ='Attributes' ,y ='Value' ,data = comb_df , hue = 'Users',errwidth=1.2)\n",
    "                ax.set_xlabel('')\n",
    "                ax.set_ylabel('')\n",
    "                plt.xticks(rotation='vertical')\n",
    "                ax.legend_.remove()\n",
    "                plt.show()\n",
    "                hate_df = pd.DataFrame()\n",
    "                counter_df = pd.DataFrame()\n",
    "                Noncounter_df = pd.DataFrame()\n",
    "            else:\n",
    "                comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "                comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "                plt.figure(figsize = (3,3))\n",
    "                ax = sns.barplot(x ='Attributes' ,y ='Value' ,data = comb_df , hue = 'Users',errwidth=1.2)\n",
    "\n",
    "                ax.set_xlabel('')\n",
    "                ax.set_ylabel('')\n",
    "                plt.xticks(rotation='vertical')\n",
    "                ax.legend_.remove()\n",
    "                plt.show()\n",
    "                hate_df = pd.DataFrame()\n",
    "                counter_df = pd.DataFrame()\n",
    "                Noncounter_df = pd.DataFrame()\n",
    "        except:\n",
    "            pass\n",
    "#################################### END PLOT ###############################################\n",
    "comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "plt.figure(figsize = (2.7,3))\n",
    "ax = sns.barplot(x ='Attributes' ,y ='Value' ,data = comb_df , hue = 'Users',errwidth=1.2)\n",
    "ax.legend_.remove()\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('')\n",
    "plt.xticks(rotation='vertical')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:33.446377Z",
     "start_time": "2018-10-26T20:19:33.442801Z"
    }
   },
   "outputs": [],
   "source": [
    "keys = ['_r_personality_parent_Agreeableness',\n",
    "'_r_personality_parent_Agreeableness_Altruism',\n",
    "'_r_personality_parent_Agreeableness_Cooperation',\n",
    "'_r_personality_parent_Agreeableness_Modesty',\n",
    "'_r_personality_parent_Agreeableness_Morality',\n",
    "'_r_personality_parent_Agreeableness_Sympathy',\n",
    "'_r_personality_parent_Agreeableness_Trust',\n",
    "'_r_personality_parent_Neuroticism_Immoderation',\n",
    "'_r_personality_parent_Neuroticism_Self-consciousness']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:34.505920Z",
     "start_time": "2018-10-26T20:19:33.448884Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 201.6x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(palette = 'muted')\n",
    "hate_df = pd.DataFrame()\n",
    "counter_df = pd.DataFrame()\n",
    "Noncounter_df = pd.DataFrame()\n",
    "for key,value in personality_output.items():\n",
    "    if key in keys:\n",
    "        temp_h = pd.DataFrame(np.c_[['hate' for i in range(len(list(personality_output.values())[0]['hate']))]]\n",
    "                                        ,columns =['Users'])\n",
    "        temp_c = pd.DataFrame(np.c_[['counter' for i in range(len(list(personality_output.values())[0]['counter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "        temp_nc = pd.DataFrame(np.c_[['Noncounter' for i in range(len(list(personality_output.values())[0]['noncounter']))]]\n",
    "                                ,columns =['Users'] )\n",
    "        temp_h['Value'],temp_c['Value'],temp_nc['Value'] = value['hate'],value['counter'],value['noncounter']\n",
    "        if len(re.findall(r'[^_]+',key)) == 3:\n",
    "            temp_h['Attributes'],temp_c['Attributes'],temp_nc['Attributes'] = re.findall(r'[^_]+',key)[-2],re.findall(r'[^_]+',key)[-2],re.findall(r'[^_]+',key)[-2]\n",
    "        else:\n",
    "            temp_h['Attributes'],temp_c['Attributes'],temp_nc['Attributes'] = re.findall(r'[^_]+',key)[-1],re.findall(r'[^_]+',key)[-1],re.findall(r'[^_]+',key)[-1]\n",
    "        hate_df = pd.concat([hate_df,temp_h],ignore_index = True)\n",
    "        counter_df = pd.concat([counter_df,temp_c],ignore_index = True)\n",
    "        Noncounter_df = pd.concat([Noncounter_df,temp_nc],ignore_index = True)\n",
    "\n",
    "comb_df = pd.concat([Noncounter_df,counter_df,hate_df],ignore_index = True)\n",
    "comb_df = comb_df.apply(pd.to_numeric ,errors = 'ignore')\n",
    "plt.figure(figsize = (2.8,3))\n",
    "ax = sns.barplot(x ='Attributes' ,y ='Value' ,data = comb_df , hue = 'Users',errwidth=1.2)\n",
    "ax.legend_.remove()\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('')\n",
    "plt.ylim(ymin = 0.06)\n",
    "title = re.findall(r'[^_]+',keys[0])[-1]\n",
    "if len(re.findall(r'[^_]+',keys[0])) == 3:\n",
    "    plt.title(re.findall(r'[^_]+',keys[0])[-2])\n",
    "else:\n",
    "    plt.title(re.findall(r'[^_]+',keys[0])[-1])\n",
    "plt.xticks(rotation='vertical')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:34.513794Z",
     "start_time": "2018-10-26T20:19:34.508278Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.09212864619707506"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.median(list(personality_output.values())[0]['hate'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:19:42.621731Z",
     "start_time": "2018-10-26T20:19:42.613509Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:3: UserWarning: \n",
      "This call to matplotlib.use() has no effect because the backend has already\n",
      "been chosen; matplotlib.use() must be called *before* pylab, matplotlib.pyplot,\n",
      "or matplotlib.backends is imported for the first time.\n",
      "\n",
      "The backend was *originally* set to 'module://ipykernel.pylab.backend_inline' by the following code:\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/runpy.py\", line 193, in _run_module_as_main\n",
      "    \"__main__\", mod_spec)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/runpy.py\", line 85, in _run_code\n",
      "    exec(code, run_globals)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py\", line 16, in <module>\n",
      "    app.launch_new_instance()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/traitlets/config/application.py\", line 658, in launch_instance\n",
      "    app.start()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/kernelapp.py\", line 486, in start\n",
      "    self.io_loop.start()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/tornado/platform/asyncio.py\", line 127, in start\n",
      "    self.asyncio_loop.run_forever()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/asyncio/base_events.py\", line 422, in run_forever\n",
      "    self._run_once()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/asyncio/base_events.py\", line 1432, in _run_once\n",
      "    handle._run()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/asyncio/events.py\", line 145, in _run\n",
      "    self._callback(*self._args)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/tornado/platform/asyncio.py\", line 117, in _handle_events\n",
      "    handler_func(fileobj, events)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 276, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 450, in _handle_events\n",
      "    self._handle_recv()\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 480, in _handle_recv\n",
      "    self._run_callback(callback, msg)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 432, in _run_callback\n",
      "    callback(*args, **kwargs)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 276, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 283, in dispatcher\n",
      "    return self.dispatch_shell(stream, msg)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 233, in dispatch_shell\n",
      "    handler(stream, idents, msg)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 399, in execute_request\n",
      "    user_expressions, allow_stdin)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/ipkernel.py\", line 208, in do_execute\n",
      "    res = shell.run_cell(code, store_history=store_history, silent=silent)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/ipykernel/zmqshell.py\", line 537, in run_cell\n",
      "    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2662, in run_cell\n",
      "    raw_cell, store_history, silent, shell_futures)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2785, in _run_cell\n",
      "    interactivity=interactivity, compiler=compiler, result=result)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2903, in run_ast_nodes\n",
      "    if self.run_code(code, result):\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2963, in run_code\n",
      "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
      "  File \"<ipython-input-9-cbd938e83622>\", line 1, in <module>\n",
      "    get_ipython().run_line_magic('matplotlib', 'inline')\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2131, in run_line_magic\n",
      "    result = fn(*args,**kwargs)\n",
      "  File \"<decorator-gen-107>\", line 2, in matplotlib\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/magic.py\", line 187, in <lambda>\n",
      "    call = lambda f, *a, **k: f(*a, **k)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/magics/pylab.py\", line 99, in matplotlib\n",
      "    gui, backend = self.shell.enable_matplotlib(args.gui)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 3051, in enable_matplotlib\n",
      "    pt.activate_matplotlib(backend)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/IPython/core/pylabtools.py\", line 311, in activate_matplotlib\n",
      "    matplotlib.pyplot.switch_backend(backend)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/matplotlib/pyplot.py\", line 231, in switch_backend\n",
      "    matplotlib.use(newbackend, warn=False, force=True)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/matplotlib/__init__.py\", line 1410, in use\n",
      "    reload(sys.modules['matplotlib.backends'])\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/importlib/__init__.py\", line 166, in reload\n",
      "    _bootstrap._exec(spec, module)\n",
      "  File \"/home/navish/anaconda3/lib/python3.6/site-packages/matplotlib/backends/__init__.py\", line 16, in <module>\n",
      "    line for line in traceback.format_stack()\n",
      "\n",
      "\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from math import pi\n",
    "matplotlib.use('Agg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-26T20:27:40.778458Z",
     "start_time": "2018-10-26T20:27:40.496709Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PolarAxesSubplot(0.260833,0.125;0.503333x0.755)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/navish/anaconda3/lib/python3.6/site-packages/matplotlib/cbook/deprecation.py:107: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n",
      "/home/navish/anaconda3/lib/python3.6/site-packages/matplotlib/cbook/deprecation.py:107: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'Challenge'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keys = ['_r_values_parent',\n",
    "_r_values_parent_Openness to change\n",
    "_r_values_parent_Hedonism\n",
    "_r_values_parent_Self-enhancement\n",
    "_r_values_parent_Self-transcendence]\n",
    "\n",
    "sns.set(palette = 'muted')\n",
    "hate_df = {}\n",
    "counter_df = {}\n",
    "Noncounter_df = {}\n",
    "for key,value in personality_output.items():\n",
    "    if key in keys:\n",
    "        if len(re.findall(r'[^_]+',key)) == 3:\n",
    "            hate_df.update({re.findall(r'[^_]+',key)[-2]:np.median(value['hate'])})\n",
    "            counter_df.update({re.findall(r'[^_]+',key)[-2]:np.median(value['counter'])})\n",
    "            Noncounter_df.update({re.findall(r'[^_]+',key)[-2]:np.median(value['noncounter'])})\n",
    "        else:\n",
    "            hate_df.update({re.findall(r'[^_]+',key)[-1]:np.median(value['hate'])})\n",
    "            counter_df.update({re.findall(r'[^_]+',key)[-1]:np.median(value['counter'])})\n",
    "            Noncounter_df.update({re.findall(r'[^_]+',key)[-1]:np.median(value['noncounter'])})\n",
    "df = pd.DataFrame([Noncounter_df,counter_df,hate_df])\n",
    "     \n",
    "skills = list(df)\n",
    "num_skills = len(skills)\n",
    "angles = [i / float(num_skills) * 2 * pi for i in range(num_skills)]\n",
    "angles += angles[:1]  # repeat the first value to close the circle\n",
    "GRAY = '#999999'\n",
    "\n",
    "# Clear the plot to start with a blank canvas.\n",
    "plt.clf()\n",
    "\n",
    "# Create subplots for each data series\n",
    "counter = plt.subplot(111,projection='polar') \n",
    "counter.set_xticks(angles)\n",
    "counter.set_xticklabels(skills, fontsize=10)\n",
    "Noncounter = plt.subplot(111,projection='polar')\n",
    "Noncounter.set_xticks(angles)\n",
    "Noncounter.set_xticklabels(skills, fontsize=10)\n",
    "hate = plt.subplot(111,projection='polar') \n",
    "hate.set_xticks(angles)\n",
    "hate.set_xticklabels(skills, fontsize=10)\n",
    "\n",
    "#######################################################################################################################\n",
    "plt.xticks(angles, skills, size=10)\n",
    "plt.yticks(np.arange(0,0.82,0.2),\n",
    "  ['0.0','0.2', '0.4', '0.6','0.8'],size=8)\n",
    "#######################################################################################################################\n",
    "# Constrain y axis to range 0-100\n",
    "# plt.ylim(ymin=0.0, ymax=0.62)\n",
    "\n",
    "Noncounter_values = df.loc[0] \\\n",
    "                    .values \\\n",
    "                    .flatten() \\\n",
    "                    .tolist()\n",
    "Noncounter_values += Noncounter_values[:1]  # duplicate first element to close the circle\n",
    "\n",
    "counter_values = df.loc[1] \\\n",
    "                    .values \\\n",
    "                    .flatten() \\\n",
    "                    .tolist()\n",
    "counter_values += counter_values[:1]   # duplicate first element to close the circle\n",
    "\n",
    "hate_values = df.loc[2] \\\n",
    "                    .values \\\n",
    "                    .flatten() \\\n",
    "                    .tolist()\n",
    "hate_values += hate_values[:1]   # duplicate first element to close the circle\n",
    "\n",
    "# Set up colors\n",
    "red = '#c44e52'\n",
    "green = '#55a868'\n",
    "blue = '#4c72b0'\n",
    "\n",
    "\n",
    "\n",
    "############################################ Plot Noncounter ########################\n",
    "# Noncounter.set_rlabel_position(0)\n",
    "# Noncounter.plot(\n",
    "#   angles,\n",
    "#   Noncounter_values,\n",
    "#   color=blue,\n",
    "#   linestyle='solid',\n",
    "#   linewidth=2,\n",
    "# )\n",
    "# Noncounter.fill(\n",
    "#   angles,\n",
    "#   Noncounter_values,\n",
    "#   color=blue,\n",
    "#   alpha=.99\n",
    "# )\n",
    "# Noncounter.set_xticks(angles)\n",
    "# Noncounter.set_xticklabels(skills, fontsize=10)\n",
    "############################################# Plot hate###########################\n",
    "hate.set_rlabel_position(0)\n",
    "hate.plot(\n",
    "  angles,\n",
    "  hate_values,\n",
    "  color=red,\n",
    "  linestyle='solid',\n",
    "  linewidth=1,\n",
    ")\n",
    "\n",
    "hate.fill(\n",
    "  angles,\n",
    "  hate_values,\n",
    "  color=red,\n",
    "  alpha=0.99\n",
    ")\n",
    "\n",
    "################################## Plot  counter ############################\n",
    "counter.set_rlabel_position(0)\n",
    "counter.plot(\n",
    "  angles,\n",
    "  counter_values,\n",
    "  color=green,\n",
    "  linestyle='solid',\n",
    "  linewidth=1\n",
    ")\n",
    "counter.fill(\n",
    "  angles,\n",
    "  counter_values,\n",
    "  color=green,\n",
    "  alpha=.99\n",
    ")\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "plt.gcf().canvas.draw()\n",
    "angles = np.linspace(0,2*np.pi,len(hate.get_xticklabels()))\n",
    "angles[np.sin(angles) > 0] = angles[np.sin(angles) > 0] + np.pi\n",
    "for ax in [Noncounter]:\n",
    "#     labels = []\n",
    "    print(ax)\n",
    "    for label, angle in zip(ax.get_xticklabels(), np.rad2deg(angles)):\n",
    "#         print(label)\n",
    "        x,y = label.get_position()\n",
    "        \n",
    "        lab = ax.text(x,y-0.03, label.get_text(), transform=label.get_transform(),\n",
    "                      ha=label.get_ha(), va=label.get_va())\n",
    "#         print(x,y)\n",
    "        lab.set_rotation(angle+90)\n",
    "#         labels.append(lab)\n",
    "    ax.set_xticklabels([])\n",
    "# plt.tight_layout()\n",
    "plt.savefig('needs.eps',format = 'eps',dpi = 1200)\n",
    "plt.show()\n",
    "skills[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}