example.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7d9f1884",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from src.contact_matrices import *"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79b54557",
   "metadata": {},
   "source": [
    "# Instructions\n",
    "\n",
    "The function `LoadMatrices` takes the contact matrices aggregated by household and deployment and outputs a dictionary of classes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c54c2207",
   "metadata": {},
   "outputs": [],
   "source": [
    "CM = LoadMatrices()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c9a0448",
   "metadata": {},
   "source": [
    "`CM` is a dictionary. To access its entries we simply type `CM[h,d]` where `h` refers to the household and `d` to the deplyoment. Households identifiers have been de-anonymized and randomized and are numbers between $0$ and $59$. Deployments instead are numbers between $0$ and $2$ and correspond to the three measurement waves.\n",
    "\n",
    "`CM[h,d]` is a *class* storing several contact matrices that can be accessed as follows:\n",
    "\n",
    "```python\n",
    "CM[h,d].C_counts\n",
    "CM[h,d].C_sec\n",
    "CM[h,d].T\n",
    "CM[h,d].P\n",
    "```\n",
    "\n",
    "The name of these matrices is consistent with the notation adopted in the paper. We now provide an example, plotting the matrix $C_{\\rm counts}$ of household $7$ in the third deployment (note that we start counting from zero)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5838092f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "h = 7\n",
    "d = 2\n",
    "\n",
    "plt.imshow(CM[h,d].C_sec, cmap = 'Blues', origin = 'lower')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b6f07d4",
   "metadata": {},
   "source": [
    "If we want to aggregate all measurements referring to a given deployment, we simply need to do as follows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "92d7fdea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "d = 2\n",
    "\n",
    "C = np.mean([CM[h,d].C_counts for h in range(60)], axis = 0)\n",
    "T = np.mean([CM[h,d].T for h in range(60)], axis = 0)\n",
    "\n",
    "R = C/T\n",
    "R = R/np.mean(R)\n",
    "plt.imshow(np.log10(R), cmap = 'PiYG', vmin = -1., vmax = 1., origin = 'lower')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "230d52bf",
   "metadata": {},
   "source": [
    "Finally, if we want to export the matrix in `.csv` format, we simply run the following command\n",
    "\n",
    "```python\n",
    "pd.DataFrame(R).to_csv('CM.csv')\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c50efbe",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}