generate_figures.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generate Histogram Figure for GREC Length Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(\"../src\")\n",
    "import config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils.grec import get_metrics\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from matplotlib import colors\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Set global plt params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "matplotlib.rcParams['font.family'] = 'Avenir'\n",
    "plt.rcParams['font.size'] = 18\n",
    "plt.rcParams['axes.linewidth'] = 2\n",
    "plt.rcParams['font.family'] = 'serif'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GREC METRICS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "relation = 'institution'\n",
    "j_file = f'{config.JSON_DIR}/{relation}_augment-200526.json'\n",
    "metrics_i = get_metrics(j_file, relation)\n",
    "\n",
    "relation = 'education'\n",
    "j_file = f'{config.JSON_DIR}/{relation}_augment-200526.json'\n",
    "metrics_ed = get_metrics(j_file, relation)\n",
    "\n",
    "relation = 'dob'\n",
    "j_file = f'{config.JSON_DIR}/{relation}_augment-200526.json'\n",
    "metrics_dob = get_metrics(j_file, relation)\n",
    "\n",
    "relation = 'pob'\n",
    "j_file = f'{config.JSON_DIR}/{relation}_augment-200526.json'\n",
    "metrics_pob = get_metrics(j_file, relation)\n",
    "\n",
    "relation = 'pod'\n",
    "j_file = f'{config.JSON_DIR}/{relation}_augment-200526.json'\n",
    "metrics_pod = get_metrics(j_file, relation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x180 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_i = metrics_i['lengths']\n",
    "min_x_i = metrics_i['min_no_outlier']\n",
    "max_x_i = metrics_i['max_no_outlier']\n",
    "\n",
    "x_ed = metrics_ed['lengths']\n",
    "min_x_ed = metrics_ed['min_no_outlier']\n",
    "max_x_ed = metrics_ed['max_no_outlier']\n",
    "\n",
    "x_dob = metrics_dob['lengths']\n",
    "min_x_dob = metrics_dob['min_no_outlier']\n",
    "max_x_dob = metrics_dob['max_no_outlier']\n",
    "\n",
    "x_pob = metrics_pob['lengths']\n",
    "min_x_pob = metrics_pob['min_no_outlier']\n",
    "max_x_pob = metrics_pob['max_no_outlier']\n",
    "\n",
    "x_pod = metrics_pod['lengths']\n",
    "min_x_pod = metrics_pod['min_no_outlier']\n",
    "max_x_pod = metrics_pod['max_no_outlier']\n",
    "\n",
    "plt.rcParams[\"font.size\"] = \"12\"\n",
    "bins = list(range(0,300,10))\n",
    "\n",
    "# full image\n",
    "fig, axs = plt.subplots(1, 5)\n",
    "\n",
    "axs[0].hist(np.clip(x_i, bins[0], bins[-1]), bins=bins, color=\"white\", edgecolor='gray', alpha=0.75)\n",
    "axs[1].hist(np.clip(x_ed, bins[0], bins[-1]), bins=bins, color=\"white\", edgecolor='gray', alpha=0.75)\n",
    "axs[2].hist(np.clip(x_dob, bins[0], bins[-1]), bins=bins, color='white', edgecolor='gray', alpha=0.75)\n",
    "axs[3].hist(np.clip(x_pob, bins[0], bins[-1]), bins=bins, color='white', edgecolor='gray', alpha=0.75)\n",
    "axs[4].hist(np.clip(x_pod, bins[0], bins[-1]), bins=bins, color='white', edgecolor='gray', alpha=0.75)\n",
    "\n",
    "\n",
    "axs[0].axvline(min_x_i, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(min_x_i))\n",
    "axs[0].axvline(x_i.mean(), color='red', linestyle='solid', linewidth=2).set_label('{:.0f}'.format(x_i.mean()))\n",
    "axs[0].axvline(max_x_i, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(max_x_i))\n",
    "\n",
    "axs[1].axvline(min_x_ed, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(min_x_ed))\n",
    "axs[1].axvline(x_ed.mean(), color='red', linestyle='solid', linewidth=2).set_label('{:.0f}'.format(x_ed.mean()))\n",
    "axs[1].axvline(max_x_ed, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(max_x_ed))\n",
    "\n",
    "axs[2].axvline(min_x_dob, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(min_x_dob))\n",
    "axs[2].axvline(x_dob.mean(), color='red', linestyle='solid', linewidth=2).set_label('{:.0f}'.format(x_dob.mean()))\n",
    "axs[2].axvline(max_x_dob, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(max_x_dob))\n",
    "\n",
    "axs[3].axvline(min_x_pob, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(min_x_pob))\n",
    "axs[3].axvline(x_pob.mean(), color='red', linestyle='solid', linewidth=2).set_label('{:.0f}'.format(x_pob.mean()))\n",
    "axs[3].axvline(max_x_pob, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(max_x_pob))\n",
    "\n",
    "axs[4].axvline(min_x_pod, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(min_x_pod))\n",
    "axs[4].axvline(x_pod.mean(), color='red', linestyle='solid', linewidth=2).set_label('{:.0f}'.format(x_pod.mean()))\n",
    "axs[4].axvline(max_x_pod, color='black', linestyle='dashed', linewidth=2).set_label('{:.0f}'.format(max_x_pod))\n",
    "\n",
    "axs[0].set_title('Institution', fontsize='small')\n",
    "axs[1].set_title('Education', fontsize='small')\n",
    "axs[2].set_title('Date of Birth', fontsize='small')\n",
    "axs[3].set_title('Place of Birth', fontsize='small')\n",
    "axs[4].set_title('Place of Death', fontsize='small')\n",
    "\n",
    "\n",
    "for ax in axs.flat:\n",
    "    ax.set_xlabel('Snippet Length', size='small')\n",
    "    ax.set_ylabel('Number of Snippets', size='small')\n",
    "for ax in axs.flat:\n",
    "    ax.label_outer()\n",
    "    \n",
    "plt.rc('xtick', labelsize='x-small')\n",
    "plt.rc('ytick', labelsize='x-small')\n",
    "fig_size = plt.gcf().get_size_inches() #Get current size\n",
    "sizefactor = 11/fig_size[0] #Set a zoom factor\n",
    "plt.gcf().set_size_inches(fig_size[0] * sizefactor, 2.5)\n",
    "plt.tight_layout()\n",
    "\n",
    "# Uncomment to save figure as .pdf\n",
    "# plt.savefig(output_dir + 'snippet_histogram.pdf', dpi=300, transparent=False, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GREC Metrics Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lll}\n",
      "\\toprule\n",
      "{} & Snippets before filter & Snippets after filter \\\\\n",
      "\\midrule\n",
      "Institution    &                    NaN &                   NaN \\\\\n",
      "Education      &                    NaN &                   NaN \\\\\n",
      "Date of Birth  &                    NaN &                   NaN \\\\\n",
      "Place of Birth &                    NaN &                   NaN \\\\\n",
      "Place of Death &                    NaN &                   NaN \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(index=['Institution', 'Education', 'Date of Birth', 'Place of Birth', 'Place of Death'], columns=['Snippets before filter', 'Snippets after filter'])\n",
    "b = [metrics_i['num_relations'], metrics_ed['num_relations'], metrics_dob['num_relations'], metrics_pob['num_relations'], metrics_pod['num_relations']]\n",
    "a = [metrics_i['num_relations_within'], metrics_ed['num_relations_within'], metrics_dob['num_relations_within'], metrics_pob['num_relations_within'], metrics_pod['num_relations_within']]\n",
    "print(df.to_latex())"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "re",
   "language": "python",
   "name": "re"
  },
  "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
}