Plot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import json\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pylab as pl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Unstack')"
      ]
     },
     "execution_count": 24,
     "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": [
    "blocks_5 = [0.747, -0.959, -0.96, -0.98]\n",
    "blocks_6 = [0.607, -0.98, -0.98, -0.98]\n",
    "blocks_7 = [0.407, -0.98, -0.98, -0.98]\n",
    "shuffle_top = [0.847, -0.98, -0.785, -0.885]\n",
    "trained = [0.844, 0.94, -0.822, 0.939]\n",
    "index = ['NLRL', 'NN', 'Random', 'Q Learning']\n",
    "df = pd.DataFrame({'5blocks': blocks_5,\n",
    "                   '6blocks': blocks_6,\n",
    "                   '7blocks': blocks_7,\n",
    "                   'shuffle top2': shuffle_top,\n",
    "                   'trained': trained}, index=index)\n",
    "ax = df.plot.bar(rot=0)\n",
    "pl.suptitle(\"Unstack\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Stack')"
      ]
     },
     "execution_count": 33,
     "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": [
    "blocks_5 = [0.915, -0.959, -0.948, -0.98]\n",
    "blocks_6 = [0.891, -0.98, -0.98, -0.98]\n",
    "blocks_7 = [0.868, -0.98, -0.98, -0.98]\n",
    "swap_right2 = [ 0.913, -0.98, -0.24, -0.98]\n",
    "columns = [0.897, -0.98, -0.215, -0.98]\n",
    "trained = [ 0.91, 0.94,  -0.292, 1]\n",
    "index = ['NLRL', 'NN', 'Random', 'Q Learning']\n",
    "df = pd.DataFrame({'5blocks': blocks_5,\n",
    "                   '6blocks': blocks_6,\n",
    "                   '7blocks': blocks_7,\n",
    "                   'swap right2': swap_right2,\n",
    "                   '2columns': columns,\n",
    "                   'trained': trained}, index=index)\n",
    "ax = df.plot.bar(rot=0)\n",
    "pl.suptitle(\"Stack\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'On')"
      ]
     },
     "execution_count": 30,
     "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": [
    "blocks_5 = [0.89, -0.98, -0.949, -0.98]\n",
    "blocks_6 = [0.865, -0.98, -0.975, -0.98]\n",
    "blocks_7 = [0.844, -0.98, -0.98, -0.98]\n",
    "shuffle_top = [0.912, -0.98, -0.821, -0.516]\n",
    "shuffle_middle = [0.914, -0.98, -0.853, -0.98]\n",
    "trained = [0.915, 0.92, -0.837, 0.92]\n",
    "index = ['NLRL', 'NN', 'Random', 'Q Learning']\n",
    "df = pd.DataFrame({'5blocks': blocks_5,\n",
    "                   '6blocks': blocks_6,\n",
    "                   '7blocks': blocks_7,\n",
    "                   'swap top': shuffle_top,\n",
    "                   'swap middle': shuffle_middle,\n",
    "                   'trained': trained}, index=index)\n",
    "ax = df.plot.bar(rot=0)\n",
    "pl.suptitle(\"On\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Cliff Walking')"
      ]
     },
     "execution_count": 31,
     "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": [
    "top_left = [0.749, -0.98, -1.115, -0.98]\n",
    "top_right = [0.809, -0.98, -0.966, -0.98]\n",
    "center = [0.859, -0.917, -0.952, 0.599]\n",
    "by6 = [0.841, -0.934, -1.101, -1.124]\n",
    "by7 = [0.824, -1.122, -1.107, -1.126]\n",
    "trained = [0.862, 0.877, -1.096, 0.878]\n",
    "index = ['NLRL', 'NN', 'Random', 'Q Learning']\n",
    "df = pd.DataFrame({'top left': top_left,\n",
    "                   'top right': top_right,\n",
    "                   'center': center,\n",
    "                   '6 by 6': by6,\n",
    "                   '7 by 7': by7,\n",
    "                   'trained': trained}, index=index)\n",
    "ax = df.plot.bar(rot=0)\n",
    "pl.suptitle(\"Cliff Walking\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Windy Cliff Walking')"
      ]
     },
     "execution_count": 32,
     "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": [
    "top_left = [0.726, 0.836, -1.376, -0.98]\n",
    "top_right = [0.834, 0.919, -1.089, -0.98]\n",
    "center = [0.672, 0.859, -1.082, 0.876]\n",
    "by6 = [0.345, -1.110, -0.907, -1.204]\n",
    "by7 = [0.506, -1.161, -1.077, -1.204]\n",
    "trained = [0.663, 0.649, -1.129, 0.7302]\n",
    "index = ['NLRL', 'NN', 'Random', 'Q Learning']\n",
    "df = pd.DataFrame({'top left': top_left,\n",
    "                   'top right': top_right,\n",
    "                   'center': center,\n",
    "                   '6 by 6': by6,\n",
    "                   '7 by 7': by7,\n",
    "                   'trained': trained}, index=index)\n",
    "ax = df.plot.bar(rot=0)\n",
    "pl.suptitle(\"Windy Cliff Walking\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}