SIRS_simulation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SIRS with mobility and network structure: SIMULATION SF vs ER"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We fixed the epidemic parameters in a \"critical\" point at which most of the times the epidemics does not survive and dies out after some iterations. The topology of the two networks is also kept fixed. We now explore the role of the mobility parameter p_mob. In this case simulation is done coupling a scale free network and an Erdos-Renyi network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt \n",
    "import networkx as nx\n",
    "import os\n",
    "import pandas as pd\n",
    "import time \n",
    "# to install tqdm\n",
    "#! pip install tqdm\n",
    "from tqdm import tnrange, trange\n",
    "\n",
    "# import custom module\n",
    "import SIRS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# needed only if you made some changes at SIRS.py and want to import the updated version of the module\n",
    "from importlib import reload\n",
    "reload(SIRS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#mobility and topology parameters\n",
    "N = 1000\n",
    "I_sf_init = 5\n",
    "I_er_init = 0\n",
    "mean_degree = 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "beta0 = 0.3\n",
    "gamma = 0.016\n",
    "mu = 0.15\n",
    "\n",
    "#mobility parameters\n",
    "p_mob_list = np.linspace(0.03,0.2,20)\n",
    "#p_mob_list = np.concatenate([[0],p_mob_list])\n",
    "print('Number of travellers:')\n",
    "for p in p_mob_list:\n",
    "    print('p=%.2f    N=%d'%(p, int(p*N)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "beta = beta0/mean_degree\n",
    "infection_params = dict(beta=beta, mu=mu, gamma=gamma)\n",
    "\n",
    "\n",
    "UP_directory_name = 'data_SF_ER/'+str(beta)+\"_\"+str(mu)+\"_\"+str(gamma) \n",
    "os.system(\"mkdir data_SF_ER\")\n",
    "os.system(\"mkdir \"+UP_directory_name)\n",
    "start = time.time()\n",
    "\n",
    "for j, p_mob in enumerate(p_mob_list):\n",
    "\n",
    "    print('Simulation %d/%d'%(j+1,len(p_mob_list)), end =' - ')\n",
    "    print(\"P_mob: %.3f\"% p_mob)\n",
    "\n",
    "    #saving stuff for future analysis\n",
    "    directory_name = UP_directory_name+\"/pmob_\"+\"{:3f}\".format(p_mob)\n",
    "    os.system(\"mkdir \"+directory_name)\n",
    "\n",
    "    #saving epidemics, topology and mobility parameters\n",
    "\n",
    "    params=dict(N=N, I_sf_init=I_sf_init, I_er_nit=I_er_init, p_mob=p_mob, mean_degree=mean_degree,\n",
    "                eps=eps, **infection_params)\n",
    "\n",
    "    parameters = pd.DataFrame(params, index=[0])\n",
    "    save=directory_name+\"/parameters.csv\"\n",
    "    parameters.to_csv(save)\n",
    "\n",
    "\n",
    "\n",
    "    #STARTING SIMULATION\n",
    "    n_runs = 100\n",
    "    n_iter = 1000\n",
    "\n",
    "\n",
    "\n",
    "    for run in tnrange(n_runs, desc='1st loop', leave=True):\n",
    "\n",
    "\n",
    "        # prepare systems\n",
    "        state_sf_init, state_er_init, variables_net_sf, variables_net_er = SIRS.prepare_two_sys(N, I_sf_init, I_er_init, \n",
    "                                                                                      p_mob, mean_degree)\n",
    "\n",
    "        #saving initial states\n",
    "        save = directory_name + \"/state_sf_init_\"+str(run)+\".txt\"\n",
    "        np.savetxt(save, state_sf_init.astype(int), fmt ='%d')\n",
    "\n",
    "        save = directory_name + \"/state_er_init_\"+str(run)+\".txt\"\n",
    "        np.savetxt(save, state_sf_init.astype(int), fmt ='%d')\n",
    "\n",
    "        save = directory_name + \"/travellers_sf_\"+str(run)+\".txt\"\n",
    "        travellers_sf = variables_net_sf[\"travellers_sf\"]\n",
    "        np.savetxt(save, travellers_sf.astype(int), fmt ='%d')\n",
    "        \n",
    "        save = directory_name + \"/travellers_er_\"+str(run)+\".txt\"\n",
    "        travellers_er = variables_net_er[\"travellers_er\"]\n",
    "        np.savetxt(save, travellers_er.astype(int), fmt ='%d')\n",
    "        \n",
    "        #saving initial variables: we just save the IDs of the travelling nodes and their degrees\n",
    "        save = directory_name + \"/deg_trav_sf_\"+str(run)+\".txt\"\n",
    "        deg_sf = variables_net_sf[\"A_sf\"].sum(axis=1)\n",
    "        deg_trav_sf = deg_sf[travellers_sf]\n",
    "        np.savetxt(save, deg_trav_sf.astype(int), fmt ='%d')\n",
    "        \n",
    "        save = directory_name + \"/deg_trav_er_\"+str(run)+\".txt\"\n",
    "        deg_er = variables_net_er[\"A_er\"].sum(axis=1)\n",
    "        deg_trav_er = deg_er[travellers_er]\n",
    "        np.savetxt(save, deg_trav_er.astype(int), fmt ='%d')\n",
    "\n",
    "        state_sf = np.copy(state_sf_init)\n",
    "        state_er = np.copy(state_er_init)\n",
    "\n",
    "        S_sf = [N - I_sf_init]\n",
    "        I_sf = [I_sf_init]\n",
    "        R_sf = [0]\n",
    "        S_er = [N - I_er_init]\n",
    "        I_er = [I_er_init]\n",
    "        R_er = [0]\n",
    "        t_vec = []\n",
    "        for i in tnrange(n_iter, desc='2nd loop', leave=False):\n",
    "            state_sf, state_er = SIRS.two_sys_full_SIRS_step(state_sf, state_er, **variables_net_sf, \n",
    "                                                    **variables_net_er, **infection_params)\n",
    "            S_sf.append(state_sf[:,0].sum())\n",
    "            I_sf.append(state_sf[:,1].sum())\n",
    "            R_sf.append(state_sf[:,2].sum())\n",
    "            S_er.append(state_er[:,0].sum())\n",
    "            I_er.append(state_er[:,1].sum())\n",
    "            R_er.append(state_er[:,2].sum())\n",
    "            #t_vec.append(time.time()-start)\n",
    "\n",
    "        #tot_time = time.time()-start\n",
    "        #print(\"Total time elapsed: %.2f s\"%tot_time)\n",
    "        #print(\"Time per iteration: %.4f s\"%(tot_time/n_iter))\n",
    "\n",
    "        S_sf = np.array(S_sf)\n",
    "        I_sf = np.array(I_sf)\n",
    "        R_sf = np.array(R_sf)\n",
    "        S_er = np.array(S_er)\n",
    "        I_er = np.array(I_er)\n",
    "        R_er = np.array(R_er)\n",
    "        t_vec = np.array(t_vec)\n",
    "\n",
    "        #save timeseries of SIR\n",
    "        SIR_sf = np.vstack((S_sf, I_sf, R_sf)).T\n",
    "        save = directory_name+\"/SIR_sf_\"+str(run)+\".txt\"\n",
    "        np.savetxt(save, SIR_sf.astype(int), fmt ='%d')\n",
    "        SIR_er = np.vstack((S_er, I_er, R_er)).T\n",
    "        save = directory_name+\"/SIR_er_\"+str(run)+\".txt\"\n",
    "        np.savetxt(save, SIR_er.astype(int), fmt ='%d')\n",
    "\n",
    "    \n",
    "    dt = (time.time() - start)/3600\n",
    "    # estimated time of arrival\n",
    "    eta = (dt/(j+1))*(len(p_mob_list)-j-1)\n",
    "    print(\"Total time elapsed: %.2f - ETA: %.2f\"%(dt, eta))"
   ]
  }
 ],
 "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": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}