Paper Charts.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "executionInfo": {
     "elapsed": 350,
     "status": "ok",
     "timestamp": 1612810795707,
     "user": {
      "displayName": "Nicholas Maltbie",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ghwu6W2KNcxJkAk73ZYWMgeRKwAP6IsD56JX9dHCiQ=s64",
      "userId": "17259711819791162155"
     },
     "user_tz": 300
    },
    "id": "PudCRdLszQ9Z"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "!mkdir -p Dataset-Analysis/charts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 58182,
     "status": "ok",
     "timestamp": 1612810853603,
     "user": {
      "displayName": "Nicholas Maltbie",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ghwu6W2KNcxJkAk73ZYWMgeRKwAP6IsD56JX9dHCiQ=s64",
      "userId": "17259711819791162155"
     },
     "user_tz": 300
    },
    "id": "RpyOYAct1rda",
    "outputId": "46a28c8c-2421-47e1-b519-3486fa5f83d0"
   },
   "outputs": [],
   "source": [
    "# ! apt-get install texlive-latex-recommended \n",
    "# ! apt install texlive-latex-extra\n",
    "# ! apt install dvipng"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 282
    },
    "executionInfo": {
     "elapsed": 975,
     "status": "ok",
     "timestamp": 1612813447706,
     "user": {
      "displayName": "Nicholas Maltbie",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ghwu6W2KNcxJkAk73ZYWMgeRKwAP6IsD56JX9dHCiQ=s64",
      "userId": "17259711819791162155"
     },
     "user_tz": 300
    },
    "id": "rrli1X240lCY",
    "outputId": "280e8917-c6be-4456-a45e-3b40c5547373"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "\n",
    "metrics = pd.DataFrame(\n",
    "    [\n",
    "     [\"LSTM\",       78.89, \"Recall Elevated\"],\n",
    "     [\"RuleMatrix\", 43.22, \"Recall Elevated\"],\n",
    "     [\"LSTM\",       41.31, \"Precision Elevated\"],\n",
    "     [\"RuleMatrix\", 17.30, \"Precision Elevated\"],\n",
    "     [\"LSTM\",       99.63, \"Recall Normal\"],\n",
    "     [\"RuleMatrix\", 99.33, \"Recall Normal\"],\n",
    "     [\"LSTM\",       99.93, \"Precision Normal\"],\n",
    "     [\"RuleMatrix\", 99.81, \"Precision Normal\"],\n",
    "    ],\n",
    "    columns=[\"Model\", \"Metric\", \"Score\"]\n",
    ")\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "sns.set_style(\"white\")\n",
    "\n",
    "matplotlib.use(\"pgf\")\n",
    "matplotlib.rcParams.update({\n",
    "    \"pgf.texsystem\": \"pdflatex\",\n",
    "    'font.family': 'serif',\n",
    "    'text.usetex': True,\n",
    "    'pgf.rcfonts': False,\n",
    "})\n",
    "\n",
    "sns.barplot(x=\"Score\", y=\"Metric\", hue=\"Model\", data = metrics)\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "plt.title(\"Performance Metrics for Predicting CSO Events\")\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "plt.savefig('Dataset-Analysis/charts/performance_metrics.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "executionInfo": {
     "elapsed": 1140,
     "status": "ok",
     "timestamp": 1612813455679,
     "user": {
      "displayName": "Nicholas Maltbie",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ghwu6W2KNcxJkAk73ZYWMgeRKwAP6IsD56JX9dHCiQ=s64",
      "userId": "17259711819791162155"
     },
     "user_tz": 300
    },
    "id": "HmBQNNXyT6E_",
    "outputId": "86c15a06-95b4-4bd1-a0d1-73acf07ec094"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 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"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "\n",
    "sig_by_feature = pd.DataFrame(\n",
    "   [[\"Flow\"         , 19.5803718,  3.1391596, 11.6318313, 18.2714378],\n",
    "    [\"Level\"        , 21.0770913,  3.2418424, 11.7781398, 18.3600123],\n",
    "    [\"Velocity\"     , 28.4718425,  3.2271735, 11.8752689, 18.3371189],\n",
    "    [\"Rainfall\"     , 30.6724986, 87.3113407, 52.9940985, 26.8062798],\n",
    "    [\"Outfall Level\", 00.1981957,  3.0804838, 11.7206615, 18.2251512],\n",
    "    ],\n",
    "    columns=[\"Feature\", \"SHAP\", \"LIME 10\", \"LIME 100\", \"LIME 720\"]\n",
    ")\n",
    "fig = plt.figure()\n",
    "fig.tight_layout()\n",
    "\n",
    "matplotlib.use(\"pgf\")\n",
    "matplotlib.rcParams.update({\n",
    "    \"pgf.texsystem\": \"pdflatex\",\n",
    "    'font.family': 'serif',\n",
    "    'text.usetex': True,\n",
    "    'pgf.rcfonts': False,\n",
    "})\n",
    "\n",
    "sns.set()\n",
    "\n",
    "sns.set_style(\"white\")\n",
    "\n",
    "sig_by_feature.set_index('Feature').T.plot(kind=\"bar\", stacked=True)\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "plt.title(\"Significance by Feature for Explanations\")\n",
    "plt.xlabel(\"XAI Tool\")\n",
    "plt.ylabel(\"Significance Percentage\")\n",
    "plt.xticks(rotation=0)\n",
    "\n",
    "plt.savefig('Dataset-Analysis/charts/significance_by_feature.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "executionInfo": {
     "elapsed": 1455,
     "status": "ok",
     "timestamp": 1612818397649,
     "user": {
      "displayName": "Nicholas Maltbie",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ghwu6W2KNcxJkAk73ZYWMgeRKwAP6IsD56JX9dHCiQ=s64",
      "userId": "17259711819791162155"
     },
     "user_tz": 300
    },
    "id": "D_w5oLwsbM4M",
    "outputId": "99c4ffc2-57ec-46b9-ab20-6ce785216b93"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 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"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "\n",
    "sig_by_feature = pd.DataFrame(\n",
    "[\n",
    "    [1 , 0.000, 0.186, 0.131, 0.093],\n",
    "    [2 , 0.000, 0.345, 0.180, 0.102],\n",
    "    [3 , 0.000, 0.208, 0.130, 0.092],\n",
    "    [4 , 0.001, 0.096, 0.101, 0.088],\n",
    "    [5 , 0.002, 0.042, 0.075, 0.083],\n",
    "    [6 , 0.006, 0.024, 0.062, 0.079],\n",
    "    [7 , 0.012, 0.021, 0.057, 0.078],\n",
    "    [8 , 0.028, 0.017, 0.054, 0.077],\n",
    "    [9 , 0.085, 0.016, 0.054, 0.078],\n",
    "    [10 , 0.162, 0.016, 0.052, 0.077],\n",
    "    [11 , 0.334, 0.015, 0.052, 0.077],\n",
    "    [12 , 0.370, 0.014, 0.050, 0.076],\n",
    "],\n",
    "    columns=[\"Hour\", \"SHAP\", \"LIME_10\", \"LIME_100\", \"LIME_720\"],\n",
    ")\n",
    "plt.figure()\n",
    "fig.tight_layout()\n",
    "\n",
    "matplotlib.use(\"pgf\")\n",
    "matplotlib.rcParams.update({\n",
    "    \"pgf.texsystem\": \"pdflatex\",\n",
    "    'font.family': 'serif',\n",
    "    'text.usetex': True,\n",
    "    'pgf.rcfonts': False,\n",
    "})\n",
    "\n",
    "sns.set()\n",
    "\n",
    "sns.set_style(\"white\")\n",
    "\n",
    "sig_by_feature.set_index('Hour').plot(kind=\"line\")\n",
    "# plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "plt.title(\"Significance by Hour for Explanations\")\n",
    "plt.xlabel(\"Hour of Input Sequence\")\n",
    "plt.ylabel(\"Significance PercentageDataset-Analysis/charts/\")\n",
    "plt.xticks(rotation=0)\n",
    "\n",
    "plt.savefig('Dataset-Analysis/charts/significance_by_hour.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "XoEKgN9uHKEj"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyOrFhi7dz/bta8+CY6dRXyU",
   "collapsed_sections": [],
   "name": "Paper Charts.ipynb",
   "provenance": []
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}