notebooks/pNEUMA-simulator.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5b8166a1",
   "metadata": {},
   "source": [
    "# pNEUMA simulator\n",
    "\n",
    "Data from: https://open-traffic.epfl.ch"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da162324",
   "metadata": {},
   "source": [
    "Optimal velocity function: \n",
    "\n",
    "$V:s \\mapsto V(s),$ \\\n",
    "$V(s) = \\max\\{0, \\ v_0 -v_0\\exp(-\\lambda v_0^{-1}(s -s_{\\text{jam}})) \\},$ \\\n",
    "$V'(s) = H(s -s_{\\text{jam}})\\lambda\\exp(-\\lambda v_0^{-1}(s -s_{\\text{jam}})),$ \\\n",
    "$V^{-1}(v) = -\\lambda^{-1}v_0 \\log(1-v_0^{-1}v)+s_{\\text{jam}},\\ 0<v<v_0.$\n",
    "\n",
    "Heterogeneous OV Model with Relaxed Noise: \n",
    "\n",
    "$\\begin{equation}\n",
    "  \\begin{cases}\n",
    "    \\dot \\varepsilon_n = -b^{-1}\\varepsilon_n +\\sigma_R\\sqrt{2 b^{-1}} \\xi_n,\\ b>0,\\\\\n",
    "    \\dot v_n = \\tau_n^{-1}[{( V_n(s_n)+\\varepsilon_n)\\times w(\\Delta t)-v_n}],\\ \\tau_n>0,\\\\\n",
    "    \\dot x_n = v_n,\n",
    "  \\end{cases}\n",
    "\\end{equation}$\n",
    "\n",
    "$\n",
    "\\begin{equation}\n",
    "  \\begin{cases}\n",
    "  \\varepsilon_n(t+\\delta) = (1-\\delta \\tilde b^{-1})\\varepsilon_n(t) + \\tilde\\sigma_R\\sqrt{\\delta 2 \\tilde b^{-1}} \\eta_n,\\ \\eta_n \\sim \\mathcal{N}(0,\\,1),\\\\\n",
    "    v_n(t+\\delta) = v_n(t) + \\delta \\frac{(V_n(s_n) +\\varepsilon_n(t+\\delta))\\times w(\\Delta t)-v_n(t)}{\\tau_n},\\\\\n",
    "    x_n(t+\\delta) = x_n(t) + \\delta v_n(t+\\delta).\n",
    "  \\end{cases}\n",
    "\\end{equation}$\n",
    "\n",
    "Anticipatory Operational Navigation: \n",
    "\n",
    "$\\begin{equation}\n",
    "  f(\\alpha)=\\min\\{d_{\\max}, v_0 \\Delta t(\\alpha)\\}\n",
    "\\end{equation}$\n",
    "$\\begin{equation}\n",
    "  d(\\alpha)^2 = d_{\\max}^2 +f(\\alpha)^2 - 2 d_{\\max}f(\\alpha)\\cos(\\alpha_0-\\alpha)\n",
    "\\end{equation}$\n",
    "$\\begin{equation}\n",
    "  \\alpha_{\\text{des}} = \\underset{}{\\mathrm{argmin}} {d(\\alpha)^2}\n",
    "\\end{equation}$\n",
    "$\\begin{equation}\n",
    "  \\dot \\alpha = \\tau^{-1}\\left[\\alpha_{\\text{des}}-\\alpha\\right]\n",
    "\\end{equation}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6968d13b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import zipfile\n",
    "import logging\n",
    "import warnings\n",
    "import itertools\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as mticker\n",
    "import matplotlib.animation as animation\n",
    "from joblib import Parallel, delayed\n",
    "from matplotlib import rcParams\n",
    "from tqdm.notebook import tqdm\n",
    "from scipy.ndimage import gaussian_filter\n",
    "from scipy.optimize import minimize\n",
    "from scipy.stats import linregress\n",
    "from numpy.polynomial import Polynomial\n",
    "from numpy import array, arange, where\n",
    "from math import exp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a1aeb172",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pNeuma_simulator import params\n",
    "from pNeuma_simulator.initialization import PoissonDisc\n",
    "from pNeuma_simulator.simulate import batch\n",
    "from pNeuma_simulator.results import normalized, loader, aggregate, percolate, zipdir, confidence_interval\n",
    "from pNeuma_simulator.animations import ring"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a4192ef3",
   "metadata": {
    "code_folding": [
     2
    ]
   },
   "outputs": [],
   "source": [
    "warnings.filterwarnings(\"ignore\")\n",
    "rcParams[\"animation.embed_limit\"] = 2**128\n",
    "rcParams.update(\n",
    "    {\n",
    "        \"font.family\": \"sans-serif\",\n",
    "        \"font.sans-serif\": \"Helvetica\",\n",
    "        \"mathtext.fontset\": \"stixsans\",\n",
    "        \"mathtext.default\": \"regular\",\n",
    "        \"font.size\": 8,\n",
    "        \"axes.titlesize\": 8,\n",
    "        \"xtick.labelsize\": 6,\n",
    "        \"ytick.labelsize\": 6,\n",
    "        \"axes.linewidth\": 0.25,\n",
    "        \"xtick.major.width\": 0.25,\n",
    "        \"ytick.major.width\": 0.25,\n",
    "    }\n",
    ")\n",
    "logging.getLogger(\"matplotlib.font_manager\").disabled = True"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efbee2b5",
   "metadata": {},
   "source": [
    "## 2D simulator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a83ad21d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1027.56x147.638 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pseudorandom number generator\n",
    "seed = 1\n",
    "rng = np.random.default_rng(seed)\n",
    "n_cars = 8\n",
    "n_moto = 8\n",
    "sampler = PoissonDisc(\n",
    "    n_cars, n_moto, cell=params.cell, L=params.L, W=params.cell * 3, k=params.k, clearance=params.clearance, rng=rng\n",
    ")\n",
    "samples, images = sampler.sample(rng)\n",
    "c_samples = samples[: 2 * n_cars]\n",
    "m_samples = samples[2 * n_cars :]\n",
    "markings = arange(-sampler.W / 2, 3 * sampler.W / 4, sampler.W / 2)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(2 * 8.7 * params.cm, 2.5 * params.cm), dpi=150)\n",
    "ax.set_aspect(\"equal\")\n",
    "for sample in samples:\n",
    "    if sample in c_samples:\n",
    "        sample.styles = {\"ec\": \"none\", \"fc\": \"k\", \"fill\": True, \"lw\": 0.6}\n",
    "    else:\n",
    "        sample.styles = {\"ec\": \"k\", \"fill\": False, \"lw\": 0.5}\n",
    "    sample.draw(ax)\n",
    "for image in images:\n",
    "    image.styles = {\"ec\": \"k\", \"fill\": False, \"lw\": 0.5, \"ls\": \"--\"}\n",
    "    image.draw(ax)\n",
    "for m_sample in rng.choice(m_samples, n_moto, replace=False):\n",
    "    m_sample.styles = {\"ec\": \"none\", \"fc\": \"k\", \"fill\": True, \"lw\": 0.5}\n",
    "    m_sample.draw(ax)\n",
    "ax.hlines(\n",
    "    [-params.lane, params.lane],\n",
    "    xmin=-sampler.width / 2,\n",
    "    xmax=sampler.width / 2,\n",
    "    color=\"k\",\n",
    "    ls=\"-\",\n",
    "    lw=1,\n",
    ")\n",
    "ax.vlines(\n",
    "    sampler.x_boundaries,\n",
    "    ymin=-3 * sampler.cell / 2,\n",
    "    ymax=3 * sampler.cell / 2,\n",
    "    color=\"k\",\n",
    "    lw=0.25,\n",
    ")\n",
    "ax.hlines(\n",
    "    sampler.y_boundaries,\n",
    "    xmin=-sampler.nx * sampler.cell / 2,\n",
    "    xmax=sampler.nx * sampler.cell / 2,\n",
    "    color=\"k\",\n",
    "    lw=0.25,\n",
    ")\n",
    "plt.xlim(-(sampler.nx + 1) * sampler.cell / 2, (sampler.nx + 1) * sampler.cell / 2)\n",
    "plt.ylim(-2 * sampler.cell, 2 * sampler.cell)\n",
    "ax.set_xlabel(\"X ($m$)\")\n",
    "ax.set_ylabel(\"Y ($m$)\")\n",
    "ax.set_yticks(arange(-4, 5, 2))\n",
    "fig.tight_layout(pad=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "98d0cc2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ID': 170,\n",
       " 'mode': 'Moto',\n",
       " 'a_des': 0,\n",
       " 'a0': 0,\n",
       " 'ttc': None,\n",
       " 'f_a': None,\n",
       " 'leader': None,\n",
       " 'gap': None,\n",
       " 'tau': None,\n",
       " 'lam': None,\n",
       " 'v0': None,\n",
       " 'd': None,\n",
       " 'pos': array([-2.55501897, -2.72772209]),\n",
       " 'vel': array([0., 0.]),\n",
       " 'theta': 0.0,\n",
       " 'interactions': []}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "samples[-1].encode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6dc347fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "56"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param = {\"const\": 4.026397, \"x1\": -0.062306}\n",
    "phi_max = int(exp(param[\"x1\"] * 0 * params.factor + param[\"const\"]))\n",
    "phi_max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0d71df9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Space discretization\n",
    "x = arange(-params.L / 2 + params.grid / 2, params.L / 2, params.grid)\n",
    "y = arange(-params.lane - params.grid / 2, params.lane + params.grid, params.grid)\n",
    "xv, yv = np.meshgrid(x, y)\n",
    "yv = np.flip(yv)\n",
    "shape = yv.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f11f23ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = \"./output/\"\n",
    "os.makedirs(path, exist_ok=True)\n",
    "# Combinatorial configurations\n",
    "n_veh = 4\n",
    "scale = 2\n",
    "l_cars = scale * arange(1, n_veh, 1)\n",
    "l_moto = scale * arange(0, n_veh - 1, 1)\n",
    "permutations = list(itertools.product(l_cars, l_moto))\n",
    "len(permutations)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d23b3a9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4e7f1f9f6c8444b0ba5d9697dbc9b63b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 0)\n",
      "(2, 2)\n",
      "(2, 4)\n",
      "(4, 0)\n",
      "(4, 2)\n",
      "(4, 4)\n",
      "(6, 0)\n",
      "(6, 2)\n",
      "(6, 4)\n"
     ]
    }
   ],
   "source": [
    "run = True\n",
    "# TODO: update code to match function in new version\n",
    "epochs = 4\n",
    "default_rng = np.random.default_rng(1024)\n",
    "seeds = default_rng.integers(1e8, size=epochs * len(permutations))\n",
    "if run:\n",
    "    with Parallel(n_jobs=-1, prefer=\"processes\") as master:\n",
    "        for n, permutation in tqdm(enumerate(permutations)):\n",
    "            start = n * epochs\n",
    "            end = (n + 1) * epochs\n",
    "            items = master(\n",
    "                delayed(batch)(seed, permutation) for seed in seeds[start:end]\n",
    "            )\n",
    "            # Dump result to a JSON data specification\n",
    "            with open(f\"{path}{permutation}.json\", \"w\") as outfile:\n",
    "              json.dump(items, outfile)\n",
    "            # Compress as Zip and delete original JSON\n",
    "            with zipfile.ZipFile(\n",
    "                f\"{path}{permutation}.zip\", \"w\", zipfile.ZIP_DEFLATED\n",
    "            ) as zipf:\n",
    "                zipdir(path, zipf)\n",
    "            print(permutation)\n",
    "            items = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bf5233a",
   "metadata": {},
   "source": [
    "# Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e4c27d69",
   "metadata": {},
   "outputs": [],
   "source": [
    "# consider only this last fraction of the simulated data\n",
    "keep = 1 / 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d9e33578",
   "metadata": {},
   "outputs": [],
   "source": [
    "# read = not run\n",
    "# path = \"./output/\"\n",
    "# os.makedirs(path, exist_ok=True)\n",
    "# if read:\n",
    "#     with open(f\"{path}results.json\", \"r\") as openfile:\n",
    "#         results = json.load(openfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d0b42c42",
   "metadata": {},
   "outputs": [],
   "source": [
    "# probe = False\n",
    "# if probe:\n",
    "#     permutation = (4, 6)\n",
    "#     items = loader(permutation, path)\n",
    "#     accidents = 0\n",
    "#     locations = []\n",
    "#     for item in items:\n",
    "#         if isinstance(item[0], list):\n",
    "#             pass\n",
    "#         else:\n",
    "#             locations.append(item)\n",
    "#             accidents += 1\n",
    "#     error = accidents / len(items)\n",
    "#     print(error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "07775312",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 0).json\n",
      "(2, 2).json\n",
      "(2, 4).json\n",
      "(4, 0).json\n",
      "(4, 2).json\n",
      "(4, 4).json\n",
      "(6, 0).json\n",
      "(6, 2).json\n",
      "(6, 4).json\n"
     ]
    }
   ],
   "source": [
    "compilation = True\n",
    "if compilation:\n",
    "    results = {}\n",
    "    for permutation in permutations:\n",
    "            n_cars, n_moto = permutation\n",
    "            results[str(permutation)] = {\n",
    "                \"VKT_cars\": [],\n",
    "                \"VHT_cars\": [],\n",
    "                \"VKT_moto\": [],\n",
    "                \"VHT_moto\": [],\n",
    "            }\n",
    "            items = loader(permutation, path)\n",
    "            accidents = 0\n",
    "            for item in items:\n",
    "                if isinstance(item[0], list):\n",
    "                    (\n",
    "                        VKT_cars,\n",
    "                        VHT_cars,\n",
    "                        VKT_moto,\n",
    "                        VHT_moto,\n",
    "                    ) = aggregate(item[0], n_cars, n_moto)\n",
    "                    results[str(permutation)][\"VKT_cars\"].append(VKT_cars)\n",
    "                    results[str(permutation)][\"VHT_cars\"].append(VHT_cars)\n",
    "                    results[str(permutation)][\"VKT_moto\"].append(VKT_moto)\n",
    "                    results[str(permutation)][\"VHT_moto\"].append(VHT_moto)\n",
    "                else:\n",
    "                    accidents += 1\n",
    "            error = accidents / len(items)\n",
    "            results[str(permutation)][\"error\"] = error\n",
    "            if n_cars > 2 and n_moto > 2:\n",
    "                percolation = percolate(items, n_moto, rng, start=5)\n",
    "            else:\n",
    "                percolation = None\n",
    "            results[str(permutation)][\"percolation\"] = percolation\n",
    "            items = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "bc0d8e11",
   "metadata": {},
   "outputs": [],
   "source": [
    "write = True\n",
    "path = \"./output/\"\n",
    "os.makedirs(path, exist_ok=True)\n",
    "if write:\n",
    "    with open(f\"{path}results.json\", \"w\") as outfile:\n",
    "        json.dump(results, outfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "62732dc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "read = not run\n",
    "path = \"./output/\"\n",
    "os.makedirs(path, exist_ok=True)\n",
    "if read:\n",
    "    with open(f\"{path}results.json\", \"r\") as openfile:\n",
    "        results = json.load(openfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2cf3b158",
   "metadata": {},
   "outputs": [],
   "source": [
    "# for permutation in permutations:\n",
    "#     str_permutation = str(permutation)\n",
    "#     print(f\"Permutation: {permutation}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "61246bc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "l_flow_cars = []\n",
    "l_flow_moto = []\n",
    "l_speed_cars = []\n",
    "l_speed_moto = []\n",
    "l_sem_cars = []\n",
    "l_sem_moto = []\n",
    "l_error = []\n",
    "for permutation in permutations:\n",
    "    flow_cars = np.nanmean(results[str(permutation)][\"VKT_cars\"]) / (\n",
    "        2e-6 * keep * params.T * params.L / params.factor\n",
    "    )\n",
    "    speed_cars = np.nanmean(results[str(permutation)][\"VKT_cars\"]) / np.nanmean(\n",
    "        results[str(permutation)][\"VHT_cars\"]\n",
    "    )\n",
    "    sem_cars = confidence_interval(results[str(permutation)][\"VKT_cars\"], rng) / np.nanmean(\n",
    "        results[str(permutation)][\"VHT_cars\"]\n",
    "    )\n",
    "    if permutation[1] > 0:\n",
    "        flow_moto = np.nanmean(results[str(permutation)][\"VKT_moto\"]) / (\n",
    "            2e-6 * keep * params.T * params.L / params.factor\n",
    "        )\n",
    "        speed_moto = np.nanmean(results[str(permutation)][\"VKT_moto\"]) / np.nanmean(\n",
    "            results[str(permutation)][\"VHT_moto\"]\n",
    "        )\n",
    "        sem_moto = confidence_interval(results[str(permutation)][\"VKT_moto\"], rng) / np.nanmean(\n",
    "            results[str(permutation)][\"VHT_moto\"]\n",
    "        )\n",
    "    else:\n",
    "        speed_moto = np.nan\n",
    "        sem_moto = np.nan\n",
    "        flow_moto = np.nan\n",
    "    error = results[str(permutation)][\"error\"]\n",
    "    l_flow_cars.append(flow_cars)\n",
    "    l_flow_moto.append(flow_moto)\n",
    "    l_speed_cars.append(speed_cars)\n",
    "    l_speed_moto.append(speed_moto)\n",
    "    l_sem_cars.append(sem_cars)\n",
    "    l_sem_moto.append(sem_moto)\n",
    "    l_error.append(error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1fad1390",
   "metadata": {},
   "outputs": [],
   "source": [
    "moto_matrix, car_matrix = np.meshgrid(arange(n_veh - 1), arange(n_veh - 1))\n",
    "car_matrix += 1\n",
    "car_matrix *= 2\n",
    "moto_matrix *= 2\n",
    "car_density = 1e3 * car_matrix / params.L\n",
    "moto_density = 1e3 * moto_matrix / (2 * params.L)\n",
    "flow_matrix_cars = array(l_flow_cars).reshape((n_veh - 1, n_veh - 1))\n",
    "flow_matrix_moto = array(l_flow_moto).reshape((n_veh - 1, n_veh - 1))\n",
    "speed_matrix_cars = array(l_speed_cars).reshape((n_veh - 1, n_veh - 1))\n",
    "speed_matrix_moto = array(l_speed_moto).reshape((n_veh - 1, n_veh - 1))\n",
    "sem_matrix_cars = array(l_sem_cars).reshape((n_veh - 1, n_veh - 1))\n",
    "sem_matrix_moto = array(l_sem_moto).reshape((n_veh - 1, n_veh - 1))\n",
    "error_matrix = array(l_error).reshape((n_veh - 1, n_veh - 1))\n",
    "percent_matrix = moto_matrix / (2 * car_matrix + moto_matrix)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "772ee7d6",
   "metadata": {},
   "source": [
    "## Rejection rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "74272eee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    error_matrix,\n",
    "    levels=arange(0.1, 1, 0.1),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels[::2], inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "aa729165",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(error_matrix.T, origin=\"lower\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5741bf4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.plot(error_matrix[-1], c=\"k\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98456d41",
   "metadata": {},
   "source": [
    "## Effect of traffic composition on the capacity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "4fa4c916",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "section = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    percent_matrix,\n",
    "    levels=[0, 0.25, 0.35],\n",
    "    colors=[\n",
    "        \"none\",\n",
    "        \"k\",\n",
    "        \"k\",\n",
    "    ],\n",
    "    linestyles=[\n",
    "        \"-\",\n",
    "        \"--\",\n",
    "        \":\",\n",
    "    ],\n",
    "    lw=1.75,\n",
    ")\n",
    "major = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    flow_matrix_cars / 1000,\n",
    "    levels=arange(0.45, 1.6, 0.05),\n",
    "    colors=\"k\",\n",
    "    frameon=True,\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.clabel(major, major.levels[3::2][4:], inline=True, fontsize=6)\n",
    "minor = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    flow_matrix_cars,\n",
    "    levels=arange(450, 1600, 1),\n",
    "    colors=\"none\",\n",
    "    frameon=True,\n",
    "    linewidths=0.5,\n",
    ")\n",
    "l_points, l_response = normalized(minor, section)\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "fig.tight_layout(pad=0)\n",
    "# plt.savefig(f\"flow_surface.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6c5bdd7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(flow_matrix_cars.T, origin=\"lower\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "df0d806a",
   "metadata": {},
   "outputs": [],
   "source": [
    "l_density = [\n",
    "    78.85164336796137,\n",
    "    80.21430502104938,\n",
    "    81.76971482132372,\n",
    "    84.2952412358987,\n",
    "    121.4738292635279,\n",
    "    126.26641429146008,\n",
    "    129.25637641049116,\n",
    "    132.76735019167575,\n",
    "    136.07715448570178,\n",
    "    139.55620001819193,\n",
    "    140.61463808620402,\n",
    "    143.44068363706938,\n",
    "    147.31179201878675,\n",
    "    55.277204341493906,\n",
    "    56.350926415363226,\n",
    "    56.728340987200056,\n",
    "    59.15088612856629,\n",
    "    47.49730502323255,\n",
    "    47.88934040436746,\n",
    "    57.64671875346068,\n",
    "    58.15279998791782,\n",
    "    59.00262311679693,\n",
    "    59.839895783363254,\n",
    "    67.74543008835644,\n",
    "    85.85646285009275,\n",
    "    88.13886959677079,\n",
    "    90.813589412546,\n",
    "    92.62579527487557,\n",
    "    95.09407661086294,\n",
    "    97.18538284222981,\n",
    "    99.37074122699903,\n",
    "    101.21331224133333,\n",
    "    103.5857831721714,\n",
    "    106.07214566317205,\n",
    "    127.67537281967729,\n",
    "    148.0210927334216,\n",
    "    151.89620982152047,\n",
    "    153.15095885255977,\n",
    "    154.93836756645743,\n",
    "    67.54819433406072,\n",
    "    68.09284618743301,\n",
    "    69.15848942390676,\n",
    "    70.63181700666524,\n",
    "    88.31011766872678,\n",
    "    90.99282148135316,\n",
    "    42.866514801061754,\n",
    "    86.25373121166918,\n",
    "    105.1148969935716,\n",
    "    109.19596135736022,\n",
    "    111.82669186663368,\n",
    "    131.4579356178662,\n",
    "    135.28949074125137,\n",
    "    138.72787252345282,\n",
    "    141.26102638279002,\n",
    "    145.013847601388,\n",
    "    147.7840554293003,\n",
    "    80.14656707719867,\n",
    "    81.1281440892312,\n",
    "    88.98274159304249,\n",
    "    100.22620105526924,\n",
    "    103.35045179907183,\n",
    "    105.46557297780107,\n",
    "    106.92324136442195,\n",
    "    108.67743619761646,\n",
    "    111.28294271429696,\n",
    "    114.57934532708677,\n",
    "    118.2284520962177,\n",
    "    120.84590917898109,\n",
    "    122.44207397691194,\n",
    "    124.41116276795671,\n",
    "    125.93127139567115,\n",
    "    127.48898604602456,\n",
    "    130.3532186962445,\n",
    "    134.70215310952977,\n",
    "    137.08915451648053,\n",
    "    141.47802606239614,\n",
    "    53.708994790095524,\n",
    "    54.210816156969244,\n",
    "    42.30121005500219,\n",
    "    42.76893332724363,\n",
    "    33.847457586750686,\n",
    "    74.0699377761518,\n",
    "    109.57030073507198,\n",
    "    113.59445814058193,\n",
    "    116.00867202329493,\n",
    "    118.37550144782684,\n",
    "    120.84091986263321,\n",
    "    122.92216595962665,\n",
    "    125.92589397872716,\n",
    "    131.08368980283393,\n",
    "    136.6820472257816,\n",
    "    149.69003222401113,\n",
    "    154.67826099957523,\n",
    "    73.42414686596992,\n",
    "    75.14823533370938,\n",
    "]\n",
    "l_flow = [\n",
    "    1558.0390273248458,\n",
    "    1546.08105538972,\n",
    "    1536.423693594767,\n",
    "    1563.3005100398946,\n",
    "    1178.0788527954155,\n",
    "    1025.529308431304,\n",
    "    1034.0376697399734,\n",
    "    885.2577182971727,\n",
    "    890.7181910575447,\n",
    "    795.2277785736013,\n",
    "    715.9356849125139,\n",
    "    643.2863450707156,\n",
    "    499.85247435434826,\n",
    "    917.9660898864796,\n",
    "    881.1661454829024,\n",
    "    1457.8094140358965,\n",
    "    1455.4988469635307,\n",
    "    1174.542585524414,\n",
    "    1137.8094812741908,\n",
    "    1425.5245177454833,\n",
    "    1466.2279320556374,\n",
    "    1423.2511880505324,\n",
    "    1472.4542604569156,\n",
    "    1527.3576816256773,\n",
    "    1509.022250248795,\n",
    "    1420.965449839972,\n",
    "    1408.9813979757166,\n",
    "    1392.1987202958474,\n",
    "    1337.106557463828,\n",
    "    1331.265283872372,\n",
    "    1300.9027626034924,\n",
    "    1239.2550217286862,\n",
    "    1205.447872494521,\n",
    "    1170.0178738361903,\n",
    "    510.6959480374308,\n",
    "    663.8294493103753,\n",
    "    533.9485994827373,\n",
    "    482.81452544546846,\n",
    "    469.53793878155653,\n",
    "    1424.6828508167503,\n",
    "    1411.4746581951968,\n",
    "    1475.529256442378,\n",
    "    1472.7253957843127,\n",
    "    1552.1492717854715,\n",
    "    1522.1920864073388,\n",
    "    805.4509928448624,\n",
    "    1526.5299661625768,\n",
    "    1541.7883529109063,\n",
    "    1588.320168709319,\n",
    "    1463.9659738500616,\n",
    "    1083.740224678931,\n",
    "    1016.7227372364727,\n",
    "    891.1374825203661,\n",
    "    701.9861089717187,\n",
    "    738.3347819645631,\n",
    "    644.7278738607845,\n",
    "    1573.8428515555174,\n",
    "    1553.792234487093,\n",
    "    1563.9655503885988,\n",
    "    1482.313610175528,\n",
    "    1352.9986995190175,\n",
    "    1380.6884717577275,\n",
    "    1347.9416489935177,\n",
    "    1212.0161704967143,\n",
    "    1187.1308696110573,\n",
    "    1111.2141118304146,\n",
    "    1017.5546007675445,\n",
    "    908.1758513337813,\n",
    "    920.1712585458981,\n",
    "    874.6625674158905,\n",
    "    778.6325179924138,\n",
    "    649.1064420204723,\n",
    "    663.6895863027673,\n",
    "    538.8013127893385,\n",
    "    481.89708505657933,\n",
    "    411.47774531381936,\n",
    "    1028.6527492987625,\n",
    "    1038.2637513959137,\n",
    "    984.533554463229,\n",
    "    995.4194428748996,\n",
    "    599.0385623165865,\n",
    "    1122.3396637820515,\n",
    "    1367.9433004251628,\n",
    "    1252.9928331526405,\n",
    "    1181.3149886039596,\n",
    "    1061.7565327478894,\n",
    "    1083.8698152031468,\n",
    "    1057.95617599896,\n",
    "    1083.8083837973725,\n",
    "    1048.6588357073506,\n",
    "    877.9972057636294,\n",
    "    689.5854098524392,\n",
    "    449.87010699758247,\n",
    "    1335.067125169194,\n",
    "    1393.6702734652763,\n",
    "]\n",
    "mask = array(l_flow) > 400\n",
    "a_density = array(l_density)[mask]\n",
    "a_flow = array(l_flow)[mask]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "85d34fef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "linestyles = [\n",
    "    \"-\",\n",
    "    \"--\",\n",
    "    \":\",\n",
    "    \"-.\",\n",
    "]\n",
    "labels = [\"  0\", \"1/4\", \"1/3\"]\n",
    "for n, (points, response) in enumerate(zip(l_points, l_response)):\n",
    "    variable = []\n",
    "    for point in l_points[n]:\n",
    "        variable.append(point[0])\n",
    "    l_variable = array(variable)\n",
    "    order = np.argsort(variable)\n",
    "    variable = np.sort(variable)\n",
    "    response = array(l_response[n])[order]\n",
    "    _ = plt.plot(\n",
    "        variable,\n",
    "        response,\n",
    "        c=\"k\",\n",
    "        ls=linestyles[n],\n",
    "        label=labels[n],\n",
    "        clip_on=False,\n",
    "        lw=1.75,\n",
    "    )\n",
    "plt.scatter(\n",
    "    a_density,\n",
    "    a_flow,\n",
    "    ec=\"k\",\n",
    "    fc=\"none\",\n",
    "    s=5,\n",
    "    lw=0.25,\n",
    "    clip_on=False,\n",
    "    zorder=10,\n",
    ")\n",
    "ax.set_ylim(0, None)\n",
    "ax.set_xlim(0, None)\n",
    "ax.spines[[\"right\", \"top\"]].set_visible(False)\n",
    "ax.set_ylabel(\"Car Flow (1/h)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "plt.legend(\n",
    "    loc=(0, 0),\n",
    "    frameon=False,\n",
    "    handletextpad=0.25,\n",
    "    handlelength=1.5,\n",
    "    ncol=3,\n",
    "    columnspacing=0.5,\n",
    "    fontsize=7,\n",
    ")\n",
    "mf = mticker.ScalarFormatter(useMathText=True)\n",
    "mf.set_powerlimits((-2, 2))\n",
    "plt.gca().yaxis.set_major_formatter(mf)\n",
    "fig.tight_layout(pad=0)\n",
    "# plt.savefig(f\"flow.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3d809894",
   "metadata": {},
   "outputs": [],
   "source": [
    "sigma = 0.1\n",
    "gauss = gaussian_filter(speed_matrix_cars, [sigma, sigma], truncate=100)\n",
    "speed_matrix_cars_f = gauss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ae196a58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    speed_matrix_cars_f,\n",
    "    levels=arange(0, 40, 1),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels[::5], inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)\n",
    "# plt.savefig(f\"speed_cars.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "19b82d5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(speed_matrix_cars.T, origin=\"lower\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "64ba12d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    sem_matrix_cars,\n",
    "    levels=arange(0, 1, 0.025),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels, inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "058e9222",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(sem_matrix_cars.T, origin=\"lower\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "68ed7edf",
   "metadata": {},
   "outputs": [],
   "source": [
    "sigma = 0.5\n",
    "gauss = gaussian_filter(speed_matrix_moto[:, 1:], [sigma, sigma], truncate=100)\n",
    "speed_matrix_moto_f = np.hstack((speed_matrix_moto[:, 0][:, np.newaxis], gauss))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c3729160",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    speed_matrix_moto_f,\n",
    "    levels=arange(0, 40, 1),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels[::5], inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)  #\n",
    "# plt.savefig(f\"speed_moto.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d67d3e46",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(speed_matrix_moto_f.T, origin=\"lower\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7042442b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 236.22x236.22 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density,\n",
    "    moto_density,\n",
    "    sem_matrix_moto,\n",
    "    levels=arange(0, 2, 0.05),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(25, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels[::2], inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "35855772",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(\n",
    "#    sem_matrix_moto.T,\n",
    "#    origin=\"lower\",\n",
    "# )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b89fb966",
   "metadata": {},
   "source": [
    "## Phase transitions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "e7c83291",
   "metadata": {},
   "outputs": [],
   "source": [
    "T_max = round(exp(params.CM))\n",
    "NU = 1.096854  # theoretical exponent for directed percolation in 1D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b900831a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def loss(x, y, x_star):\n",
    "    renorm = np.abs(x - x_star)\n",
    "    slope, _, _, _, _ = linregress(np.log(renorm)[y > 0], np.log(y[y > 0]))\n",
    "    return abs(NU - slope)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "23512c38",
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit(x, y):\n",
    "    # https://stackoverflow.com/questions/3843017/\n",
    "    zero_crossings = where(np.diff(np.sign(y)))[0]\n",
    "    x0 = x[zero_crossings[0]]\n",
    "    # https://stackoverflow.com/questions/24767191/\n",
    "    res = minimize(lambda x_star: loss(x, y, x_star), method=\"Nelder-Mead\", x0=x0)\n",
    "    popt = res.x[0]\n",
    "    return popt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "68c9eda2",
   "metadata": {},
   "outputs": [],
   "source": [
    "l_T_c = []\n",
    "l_p_c = []\n",
    "l_slope = []\n",
    "r_values = []\n",
    "p_values = []\n",
    "for permutation in permutations:\n",
    "    if permutation[0] > 2 and permutation[1] > 4:\n",
    "        x, y, _, _ = results[str(permutation)][\"percolation\"]\n",
    "        x, y = array(x), array(y)\n",
    "        try:\n",
    "            T_c = fit(x, y)\n",
    "            renorm = np.abs(T_c - x)\n",
    "            slope, intercept, r_value, p_value, std_err = linregress(\n",
    "                np.log(renorm)[y > 0], np.log(y[y > 0])\n",
    "            )\n",
    "            r_values.append(r_value)\n",
    "            p_values.append(p_value)\n",
    "            p_c = 1 - T_c / T_max\n",
    "            if r_value < 0.95:\n",
    "                print(permutation, r_value)\n",
    "                T_c = np.nan\n",
    "                p_c = np.nan\n",
    "            else:\n",
    "                l_slope.append(slope)\n",
    "        except:\n",
    "            print(f\"problem{permutation}\")\n",
    "            break\n",
    "    else:\n",
    "        T_c = np.nan\n",
    "        p_c = np.nan\n",
    "    l_T_c.append(T_c)\n",
    "    l_p_c.append(p_c)\n",
    "moto_matrix, car_matrix = np.meshgrid(arange(n_veh - 1), arange(n_veh - 1))\n",
    "car_matrix += 1\n",
    "car_matrix *= 2\n",
    "moto_matrix *= 2\n",
    "percolation_matrix = array(l_p_c).reshape((n_veh - 1, n_veh - 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "134ac55f",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.mean(r_values), np.mean(p_values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "622fbc69",
   "metadata": {},
   "outputs": [],
   "source": [
    "# _ = plt.imshow(\n",
    "#    percolation_matrix.T,\n",
    "#    origin=\"lower\",\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eaea6e80",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "surface = ax.contour(\n",
    "    car_density[1:, 3:],\n",
    "    moto_density[1:, 3:],\n",
    "    percolation_matrix[1:, 3:],\n",
    "    levels=arange(0, 1, 0.025),\n",
    "    colors=\"k\",\n",
    "    linewidths=0.5,\n",
    ")\n",
    "ax.set_xticks(arange(50, 175, 25))\n",
    "ax.set_ylabel(\"Moto Density (1/km)\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.clabel(surface, surface.levels[::8], inline=True, fontsize=6)\n",
    "fig.tight_layout(pad=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1739f5a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "markers = [\"o-\", \"s-\", \"D-\", \"X-\"]\n",
    "for i in range(4):\n",
    "    ax.plot(\n",
    "        car_density[1:, i + 3],\n",
    "        percolation_matrix[1:, i + 3],\n",
    "        markers[i],\n",
    "        c=\"k\",\n",
    "        lw=0.5,\n",
    "        ms=3,\n",
    "        mfc=\"w\",\n",
    "        mew=\".75\",\n",
    "        label=f\"$N={2*(i+3)}$\",\n",
    "    )\n",
    "ax.set_ylabel(\"$p_c$\")\n",
    "ax.set_xlabel(\"Car Density (1/km)\")\n",
    "ax.legend(\n",
    "    loc=[0, 0],\n",
    "    frameon=False,\n",
    "    handletextpad=0.25,\n",
    "    labelspacing=0.3,\n",
    "    handlelength=0.75,\n",
    "    fontsize=7,\n",
    ")\n",
    "ax.text(\n",
    "    0.95,\n",
    "    0.95,\n",
    "    \"$p_c = 1-T_c/T_{max}$\",\n",
    "    ha=\"right\",\n",
    "    va=\"top\",\n",
    "    fontsize=7,\n",
    "    transform=ax.transAxes,\n",
    ")\n",
    "# ax.set_xlim(25, 175)\n",
    "ax.set_ylim(0, 0.8)\n",
    "ax.set_xticks(arange(50, 175, 25))\n",
    "ax.set_yticks(arange(0, 0.9, 0.1))\n",
    "fig.tight_layout(pad=0)\n",
    "# plt.savefig(f\"criticality.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "c7765155",
   "metadata": {},
   "outputs": [],
   "source": [
    "def draw_law(x, y, save=False):\n",
    "    fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "    _ = plt.loglog(\n",
    "        x[y > 0],\n",
    "        y[y > 0],\n",
    "        \".\",\n",
    "        mec=\"k\",\n",
    "        mfc=\"none\",\n",
    "        ms=5,\n",
    "        markeredgewidth=0.7,\n",
    "    )\n",
    "    p_fitted = Polynomial.fit(np.log(x[y > 0]), np.log(y[y > 0]), deg=1)\n",
    "    plt.plot(\n",
    "        np.exp(np.unique(np.log(x)[y > 0])),\n",
    "        np.exp(p_fitted(np.unique(np.log(x)[y > 0]))),\n",
    "        c=\"k\",\n",
    "        ls=\"-\",\n",
    "        lw=0.5,\n",
    "    )\n",
    "    ax.text(\n",
    "        0.05,\n",
    "        0.95,\n",
    "        f\"$R^2$={r_value:.3f}\",\n",
    "        ha=\"left\",\n",
    "        va=\"top\",\n",
    "        fontsize=7,\n",
    "        transform=ax.transAxes,\n",
    "    )\n",
    "    ax.text(\n",
    "        0.05,\n",
    "        0.8,\n",
    "        f\"$\\\\nu$={slope:.4f}\",\n",
    "        ha=\"left\",\n",
    "        va=\"top\",\n",
    "        fontsize=7,\n",
    "        transform=ax.transAxes,\n",
    "    )\n",
    "    ax.text(\n",
    "        0.95,\n",
    "        0.05,\n",
    "        f\"$T_c=${T_c:.2f}\",\n",
    "        ha=\"right\",\n",
    "        va=\"bottom\",\n",
    "        fontsize=7,\n",
    "        transform=ax.transAxes,\n",
    "    )\n",
    "    ax.text(\n",
    "        0.95,\n",
    "        0.2,\n",
    "        f\"{permutation}\",\n",
    "        ha=\"right\",\n",
    "        va=\"bottom\",\n",
    "        fontsize=7,\n",
    "        transform=ax.transAxes,\n",
    "    )\n",
    "    # ax.spines[[\"right\", \"top\"]].set_visible(False)\n",
    "    ax.set_ylabel(\"$\\langle \\Delta \\Phi \\\\rangle^+$\")\n",
    "    ax.set_xlabel(\"$|T-T_c|$\")\n",
    "    fig.tight_layout(pad=0)\n",
    "    if save:\n",
    "        plt.savefig(\n",
    "            # f\"./power-laws/power-law{permutation}.pdf\", format=\"pdf\", transparent=True\n",
    "            f\"./power-law{permutation}.pdf\",\n",
    "            format=\"pdf\",\n",
    "            transparent=True,\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "bbbcca36",
   "metadata": {},
   "outputs": [],
   "source": [
    "# for permutation in permutations:\n",
    "#    if permutation[0] > 2 and permutation[1] > 4:\n",
    "#        x, y, low, high = results[str(permutation)][\"percolation\"]\n",
    "#        x, y, low, high = array(x), array(y), array(low), array(high)\n",
    "#        T_c = fit(array(x), y)\n",
    "#        renorm = np.abs(T_c - x)\n",
    "#        slope, intercept, r_value, p_value, std_err = linregress(\n",
    "#            np.log(renorm)[y > 0], np.log(y[y > 0])\n",
    "#        )\n",
    "#        p_c = 1 - T_c / T_max\n",
    "#        draw_law(renorm, y, save=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "481eb679",
   "metadata": {},
   "outputs": [],
   "source": [
    "permutation = (12, 8)\n",
    "x, y, low, high = results[str(permutation)][\"percolation\"]\n",
    "x, y, low, high = array(x), array(y), array(low), array(high)\n",
    "T_c = fit(array(x), y)\n",
    "renorm = abs(T_c - x)\n",
    "slope, intercept, r_value, p_value, std_err = linregress(\n",
    "    np.log(renorm)[y > 0], np.log(y[y > 0])\n",
    ")\n",
    "p_c = 1 - T_c / T_max\n",
    "print(\n",
    "    abs(NU - slope),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86d4f16a",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(4 * params.cm, 4 * params.cm), dpi=150)\n",
    "ax.errorbar(\n",
    "    x,\n",
    "    y,\n",
    "    yerr=high - low,\n",
    "    ms=5,\n",
    "    fmt=\".\",\n",
    "    mec=\"k\",\n",
    "    mfc=\"none\",\n",
    "    mew=0.5,\n",
    "    ecolor=\"k\",\n",
    "    elinewidth=0.5,\n",
    "    label=\"$\\langle \\Delta\\Phi \\\\rangle$\",\n",
    ")\n",
    "ax.set_xlim(0, 60)\n",
    "ax.set_ylim(-0.2, 0.6)\n",
    "ax.set_xticks(arange(0, 70, 10))\n",
    "ax.set_yticks(arange(-0.2, 0.8, 0.1))\n",
    "ax.axvline(T_c, c=\"k\", lw=0.75, ls=\"--\", label=\"$T_c$\")\n",
    "\n",
    "ax.set_ylabel(\"Percolation $\\Delta\\Phi$\")\n",
    "ax.set_xlabel(\"Temperature T\")\n",
    "ax.legend(\n",
    "    loc=[0, 0],\n",
    "    frameon=False,\n",
    "    handletextpad=0.5,\n",
    "    handlelength=1.5,\n",
    "    fontsize=7,\n",
    ")\n",
    "fig.tight_layout(pad=0)\n",
    "# plt.savefig(f\"percolation.pdf\", format=\"pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "47997128",
   "metadata": {},
   "outputs": [],
   "source": [
    "draw_law(renorm, y, save=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "87df108d",
   "metadata": {},
   "source": [
    "## Animations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd298297",
   "metadata": {},
   "outputs": [],
   "source": [
    "non_empty = []\n",
    "empty = []\n",
    "items = None\n",
    "n_cars = 12\n",
    "n_moto = 8\n",
    "permutation = (n_cars, n_moto)\n",
    "n_cars, n_moto = permutation\n",
    "load = True\n",
    "if load:\n",
    "    items = loader(permutation)\n",
    "for item in items:\n",
    "    if isinstance(item[0], list):\n",
    "        non_empty.append(item)\n",
    "    else:\n",
    "        empty.append(item)\n",
    "len(non_empty)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5dcb6fb0",
   "metadata": {},
   "outputs": [],
   "source": [
    "item = non_empty[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c76dd02c",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(14.7 * params.cm, 2.5 * params.cm), dpi=150)\n",
    "fig.tight_layout(pad=0)\n",
    "ax.set_aspect(\"equal\")\n",
    "# frame = 1\n",
    "# if frame < COUNT -1:\n",
    "#    frame += 1\n",
    "# else:\n",
    "#    frame = 1\n",
    "frame = -1\n",
    "ring(frame, item[0], ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "90e24b10",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(14.7 * params.cm, 2.5 * params.cm), dpi=150)\n",
    "fig.tight_layout(pad=0)\n",
    "ax.set_aspect(\"equal\")\n",
    "\n",
    "\n",
    "def animate_ring(i, ax=ax):\n",
    "    ax.clear()\n",
    "    ring(i, item[0], ax)\n",
    "\n",
    "\n",
    "# Create the animation\n",
    "ani_ring = animation.FuncAnimation(fig, animate_ring, frames=range(1, len(item[0]), 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4b8851ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the animation as an HTML video\n",
    "ani_ring_html = ani_ring.to_jshtml()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8e98e76e",
   "metadata": {},
   "outputs": [],
   "source": [
    "file = r\"ring.gif\"\n",
    "writergif = animation.PillowWriter(fps=1 / params.dt)\n",
    "ani_ring.save(file, writer=writergif)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dc303ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "def navi(t, l_agents, ID, ax):\n",
    "    a_des = l_agents[t][ID - 1][\"a_des\"]\n",
    "    a0 = l_agents[t][ID - 1][\"a0\"]\n",
    "    alphas = l_agents[t][ID - 1][\"alphas\"]\n",
    "    f_a = l_agents[t][ID - 1][\"f_a\"]\n",
    "    ax.invert_xaxis()\n",
    "    ax.set_yticks([])\n",
    "    ax.plot(np.degrees(alphas), f_a, c=\"k\", clip_on=False)\n",
    "    ax.scatter(np.degrees(a_des), 0, clip_on=False, marker=\"+\", c=\"k\")\n",
    "    ax.spines[\"left\"].set_position(\"zero\")\n",
    "    ax.set_xlabel(\"Direction $α$ ($\\deg$)\")\n",
    "    ax.axhline(params.d_max, color=\"k\", ls=\"--\")\n",
    "    ax.axvline(np.degrees(a0), color=\"k\", ls=\":\")\n",
    "    ax.text(np.degrees(a0) + 2, params.d_max + 0.5, \"$α_0$\")\n",
    "    ax.set_ylim(0, params.d_max)\n",
    "    ax.spines[[\"top\", \"right\"]].set_visible(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "31792727",
   "metadata": {},
   "outputs": [],
   "source": [
    "ID = 2 * n_cars + n_moto\n",
    "ID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "78664146",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(6 * params.cm, 5 * params.cm), dpi=150)\n",
    "fig.tight_layout(pad=0)\n",
    "# frame = 1\n",
    "# if frame < COUNT -1:\n",
    "#    frame += 1\n",
    "# else:\n",
    "#    frame = 1\n",
    "frame = -1\n",
    "navi(frame, item[0], ID, ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13b23a86",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(6 * params.cm, 5 * params.cm), dpi=150)\n",
    "\n",
    "\n",
    "def animate_navi(i, ax=ax):\n",
    "    ax.clear()\n",
    "    ax = navi(i, item[0], ID, ax)\n",
    "\n",
    "\n",
    "# Create the animation\n",
    "ani_navi = animation.FuncAnimation(fig, animate_navi, frames=range(1, len(item[0]), 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ab7f0a57",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the animation as an HTML video\n",
    "ani_navi_html = ani_navi.to_jshtml()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5c95de56",
   "metadata": {},
   "outputs": [],
   "source": [
    "f = r\"navi.gif\"\n",
    "writergif = animation.PillowWriter(fps=1 / params.dt)\n",
    "ani_navi.save(f, writer=writergif)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pneuma-simulator",
   "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}