Example System - Hanson 2020.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Notebook to calculate the spectrum of $\\Phi$ values for Hanson and Walker 2020 (Binary Counter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Computing concepts:   0%|          | 0/7 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Network =  NodeLabels(('Q1', 'Q2', 'Q3'))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PyPhi Value =  1.758333\n",
      "\tNumber of Non-unique Constellations = 2592\n",
      "\n",
      "Evaluating Cut  Cut [Q1] ━━/ /━━➤ [Q2, Q3]\n",
      "\tNumber of Non-unique Constellations = 18\n",
      "\n",
      "Evaluating Cut  Cut [Q2] ━━/ /━━➤ [Q1, Q3]\n",
      "\tNumber of Non-unique Constellations = 24\n",
      "\n",
      "Evaluating Cut  Cut [Q1, Q2] ━━/ /━━➤ [Q3]\n",
      "\tNumber of Non-unique Constellations = 288\n",
      "\n",
      "Evaluating Cut  Cut [Q3] ━━/ /━━➤ [Q1, Q2]\n"
     ]
    }
   ],
   "source": [
    "import pyphi\n",
    "import caffeine\n",
    "import numpy as np\n",
    "from pyphi import phi_spectrum\n",
    "\n",
    "caffeine.on(display=True)\n",
    "\n",
    "# Transition probability matrix used in Hanson and Walker 2020 Figure 4\n",
    "\n",
    "## Little End TPM. \n",
    "tpm = np.array([\n",
    "    [1,1,0],\n",
    "    [0,0,0],\n",
    "    [1,0,1],\n",
    "    [0,1,0],\n",
    "    [1,0,0],\n",
    "    [1,1,1],\n",
    "    [0,0,1],\n",
    "    [0,1,1]\n",
    "])\n",
    "\n",
    "# Set up network object\n",
    "network = pyphi.Network(tpm, node_labels=['Q1','Q2','Q3'])\n",
    "print(\"Network = \",network.node_labels)\n",
    "\n",
    "# Put the system into a given state\n",
    "state = (1,1,1)\n",
    "nodes = ['Q1','Q2','Q3']\n",
    "\n",
    "## Get the requisite Subsystem\n",
    "subsystem = pyphi.Subsystem(network, state, nodes)\n",
    "\n",
    "## What does PyPhi Give? Should be 0.125\n",
    "pyphi_val = pyphi.compute.phi(subsystem)\n",
    "print(\"PyPhi Value = \",pyphi_val)\n",
    "\n",
    "## Calculate all Phi values\n",
    "display_CES= False  # if True, output will display constellations\n",
    "Phi_Spectrum = phi_spectrum.get_phi_spectrum(subsystem,display_CES)\n",
    "\n",
    "print(\"\\nCuts = \",Phi_Spectrum[0])\n",
    "print(\"\\nPhi Spectrum = \",Phi_Spectrum[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "Phi_MIP = phi_spectrum.get_Phi_MIP(Phi_Spectrum)\n",
    "print(\"Phi MIP = \",Phi_MIP)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fontTools.ttLib import TTFont\n",
    "import matplotlib.font_manager as mfm\n",
    "\n",
    "## Choose a font that renders the cut symbol properly\n",
    "def char_in_font(Unicode_char, font):\n",
    "    for cmap in font['cmap'].tables:\n",
    "        if cmap.isUnicode():\n",
    "            if ord(Unicode_char) in cmap.cmap:\n",
    "                return True\n",
    "    return False\n",
    "\n",
    "\n",
    "uni_char = \"\\u27A4\"\n",
    "\n",
    "font_info = [(f.fname, f.name) for f in mfm.fontManager.ttflist]\n",
    "for i, font in enumerate(font_info):\n",
    "    if char_in_font(uni_char, TTFont(font[0], fontNumber=0)):\n",
    "        print(font[0], font[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from pyphi import models\n",
    "sns.set_style('darkgrid')\n",
    "%matplotlib inline\n",
    "\n",
    "## UTF Characters\n",
    "LINE = \"\\u2501\"\n",
    "CUT_SYMBOL = LINE * 2 + \"/ /\" + LINE * 2 + \"\\u27A4\"\n",
    "\n",
    "## Set Font\n",
    "# font_path = '/usr/local/lib/python3.6/site-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf'\n",
    "font_path = '/Users/jakehanson/.brew/lib/python3.7/site-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf'\n",
    "font = mfm.FontProperties(fname=font_path) # find this font\n",
    "\n",
    "## Initialize fig, ax objects\n",
    "fig = plt.figure(figsize=(8.8,5.4))\n",
    "ax = plt.subplot(111)\n",
    "\n",
    "## Plot published value and possible values\n",
    "plt.plot([0],pyphi_val,'kx',label=r'Published Value')\n",
    "ax.axhspan(np.min(Phi_MIP), np.max(Phi_MIP), alpha=0.15,color='black',label='Possible Values')\n",
    "# ax.axhspan(np.min(Phi_MIP), np.max(Phi_MIP), alpha=0.15,color='black',label='Possible $\\Phi^{Max}$ Values')\n",
    "\n",
    "## Scatterplot Phi Values for each cut\n",
    "x_labels = []\n",
    "for i in range(len(Phi_Spectrum[0])):\n",
    "    cut = Phi_Spectrum[0][i]\n",
    "    \n",
    "    ## Get label for cut\n",
    "    from_nodes = models.fmt.fmt_mechanism(cut.from_nodes, cut.node_labels)\n",
    "    to_nodes = models.fmt.fmt_mechanism(cut.to_nodes, cut.node_labels)\n",
    "    cut_label = from_nodes+CUT_SYMBOL+to_nodes\n",
    "    \n",
    "    ## Plot\n",
    "    plt.scatter([i]*len(Phi_Spectrum[1][i]),Phi_Spectrum[1][i],alpha=0.5)\n",
    "    x_labels.append(cut_label)\n",
    "    \n",
    "## Clean Up\n",
    "plt.title('Farnsworth 2021 Full System (ON)')\n",
    "plt.ylabel(r'$\\Phi$ Value(s)',fontsize=12)\n",
    "plt.xticks(range(len(Phi_Spectrum[0])),x_labels,rotation=90,fontproperties=font)\n",
    "# plt.ylim((0.0,1.0))\n",
    "plt.legend()\n",
    "# plt.legend(ncol=1,bbox_to_anchor = [1.25,1.0],prop=font,fontsize=12)\n",
    "plt.tight_layout()\n",
    "# plt.savefig(\"Farnsworth_Full_ON.pdf\")\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}