{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5af600c9-eb11-48fb-ba83-0a2c2d6e824e",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "source": [
    "(categorical_hgf)=\n",
    "# The categorical Hierarchical Gaussian Filter\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ilabcode/pyhgf/blob/master/docs/source/notebooks/1.3-CAtegorical_HGF.ipynb)\n",
    "\n",
    "```{warning}\n",
    "The categorical state node and the categorical state-transition nodes are still work in progress. The examples provided here are given for illustration. Things may change or not work until the official publication.\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e9143ab5-131e-4a1c-9130-326c3538f888",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": [
     "hide-cell"
    ]
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "from IPython.utils import io\n",
    "if 'google.colab' in sys.modules:\n",
    "\n",
    "  with io.capture_output() as captured:\n",
    "      ! pip install pyhgf watermark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d2c1e257-91ab-455d-a3ea-0ac0136797ed",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": [
     "hide-cell"
    ]
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.\n"
     ]
    }
   ],
   "source": [
    "import jax.numpy as jnp\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pytensor.tensor as pt\n",
    "import seaborn as sns\n",
    "from jax import jit, vjp\n",
    "from jax.tree_util import Partial\n",
    "from pyhgf.model import Network\n",
    "from pyhgf.plots import plot_nodes\n",
    "from pytensor.graph import Apply, Op"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fbb1fadf-bc75-4695-b62c-0db2d3d8689b",
   "metadata": {},
   "source": [
    "The binary state nodes that we introduced in the previous section are useful to encode information about stochastic boolean variables that are common in reinforcement learning and decision-making design. However, situations may occur where discrete variables can have more than two categories and therefore need to be encoded by a categorical distribution. Here, we introduce two probabilistic nodes tailored to handle this kind of variable: the **categorical state node** and the **categorical state-transition node**.\n",
    "\n",
    "Both nodes are a generalisation of the binary HGFs (in the sense that they internally represent a collection of binary state nodes). We refer to **categorical HGF** in a broad sense for HGFs that can handle categorical distributions, but as we will illustrate below, there are many ways to do that and a more precise terminology is to refer to the kind of node used internally (the **categorical state node** and the **categorical state-transition node**)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66a3991f-2170-43d6-8bb9-2541e80bf3d2",
   "metadata": {},
   "source": [
    "## Simulating a dataset\n",
    "We start by simulating a dataset on which we can apply the categorical HGFs. The dataset consists of a categorical input where the number of categories $K=10$. The underlying contingencies are generated by three Dirichlet distributions on which we sample 150 observations sequentially."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dba7fabb-e5d9-48d6-8c8e-2fecc7df28be",
   "metadata": {},
   "outputs": [],
   "source": [
    "# generate some categorical inputs data using three underlying distributions\n",
    "p1 = np.random.dirichlet(alpha=[1, 2, 3, 5, 9, 13, 17, 25, 30, 35])\n",
    "p2 = np.random.dirichlet(alpha=[1, 2, 3, 5, 30, 30, 5, 3, 2, 1])\n",
    "p3 = np.random.dirichlet(alpha=[35, 30, 25, 17, 13, 9, 5, 3, 2, 1])\n",
    "input_data = np.array(\n",
    "    [np.random.multinomial(n=1, pvals=p) for p in [p1, p2, p3] for _ in range(250)]\n",
    ").T"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5cf4027a-be50-42cf-88d7-e0847c1f9930",
   "metadata": {},
   "source": [
    "The Dirichlet distributions are parametrized in such a way that it goes from a \"skewed\" distribution to a centred distribution to another \"skewed\" distribution. The resulting sequence of categorical observations then looks like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f37e6acb-ae8e-4aea-8914-5333cbe46921",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 3))\n",
    "plt.imshow(input_data, interpolation=\"none\", aspect=\"auto\", cmap=\"binary\")\n",
    "plt.ylabel(\"Categories\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.title(\"Categorical observations\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15a23310-1d29-4a84-9cd2-a30c0e0b64e3",
   "metadata": {},
   "source": [
    "```{note}\n",
    "The lower part of the figure represent the surprise associated with the categorical node. Here, we use the [Kullback-Leibler divergence between two Dirichlet distributions](https://statproofbook.github.io/P/dir-kl.html) as a measure of Bayesian surprise. The Kullback-Leibler divergence of the Dirichlet distribution $P$ from the Dirichlet distribution $Q$ is given by the following equation:\n",
    "\n",
    "$$\n",
    "KL[P||Q] = \\ln{\\frac{\\Gamma(\\sum_{i=1}^k\\alpha_{1i})}{\\Gamma(\\sum_{i=1}^k\\alpha_{2i})}} + \\sum_{i=1}^k \\ln{\\frac{\\Gamma(\\alpha_{2i})}{\\Gamma(\\alpha_{1i})}} + \\sum_{i=1}^k(\\alpha_{1i} - \\alpha_{2i}) \\left[ \\psi(\\alpha_{1i}) - \\psi(\\sum_{i=1}^k \\alpha_{1i}) \\right]\n",
    "$$\n",
    "\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6511fd24-48a0-4913-9cb2-2ed3dba57f4f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# adding a blank input time series for the categorical state node\n",
    "# this is because the categorical state node does not receive anything\n",
    "# only binary nodes are the actual inputs of the network\n",
    "input_data = np.vstack([[0.0] * input_data.shape[1], input_data])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc71d527-2f88-4c4f-b81e-5ededd64d678",
   "metadata": {},
   "source": [
    "## The categorical state node"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19c97cda-64b8-4e1e-b90a-7bf538c033e1",
   "metadata": {},
   "source": [
    "### Creating the probabilistic network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1ae2e67a-0b7b-462b-89b7-0c1ec1fb4262",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "categorical_hgf = Network().add_nodes(\n",
    "    kind=\"categorical-input\", n_categories=10, binary_parameters={\"tonic_volatility_2\": -2.0},\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2806c2d9-e6a9-4113-829f-1f0616a31b4b",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 9.0.0 (20231125.0833)\n",
       " -->\n",
       "<!-- Title: hgf&#45;nodes Pages: 1 -->\n",
       "<svg width=\"710pt\" height=\"355pt\"\n",
       " viewBox=\"0.00 0.00 710.00 355.40\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 351.4)\">\n",
       "<title>hgf&#45;nodes</title>\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-351.4 706,-351.4 706,4 -4,4\"/>\n",
       "<!-- x_0 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>x_0</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"351,-36 303.13,-18 351,0 398.87,-18 351,-36\"/>\n",
       "<text text-anchor=\"middle\" x=\"351\" y=\"-13.32\" font-family=\"Times,serif\" font-size=\"14.00\">Ca&#45;0</text>\n",
       "</g>\n",
       "<!-- x_1 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>x_1</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"54,-108 0,-108 0,-72 54,-72 54,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"27\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;1</text>\n",
       "</g>\n",
       "<!-- x_1&#45;&gt;x_0 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>x_1&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M54.22,-75.25C57.14,-74.05 60.11,-72.93 63,-72 145.17,-45.54 245.04,-30.86 302.98,-23.98\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"303.3,-27.46 312.84,-22.84 302.5,-20.51 303.3,-27.46\"/>\n",
       "</g>\n",
       "<!-- x_2 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>x_2</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"126,-108 72,-108 72,-72 126,-72 126,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"99\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;2</text>\n",
       "</g>\n",
       "<!-- x_2&#45;&gt;x_0 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>x_2&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M126.28,-75.43C129.19,-74.19 132.13,-73.02 135,-72 193.79,-51.14 264.41,-35.5 308.51,-26.8\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"309.03,-30.27 318.18,-24.92 307.7,-23.39 309.03,-30.27\"/>\n",
       "</g>\n",
       "<!-- x_3 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>x_3</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"198,-108 144,-108 144,-72 198,-72 198,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"171\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;3</text>\n",
       "</g>\n",
       "<!-- x_3&#45;&gt;x_0 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>x_3&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M198.4,-75.75C201.29,-74.45 204.2,-73.18 207,-72 243.71,-56.61 286.62,-41.13 315.99,-30.91\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"316.8,-34.33 325.11,-27.76 314.52,-27.72 316.8,-34.33\"/>\n",
       "</g>\n",
       "<!-- x_4 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>x_4</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"270,-108 216,-108 216,-72 270,-72 270,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"243\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;4</text>\n",
       "</g>\n",
       "<!-- x_4&#45;&gt;x_0 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>x_4&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M269.97,-71.52C286.63,-60.72 307.92,-46.92 324.49,-36.18\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"326.36,-39.14 332.85,-30.76 322.55,-33.27 326.36,-39.14\"/>\n",
       "</g>\n",
       "<!-- x_5 -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>x_5</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"342,-108 288,-108 288,-72 342,-72 342,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"315\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;5</text>\n",
       "</g>\n",
       "<!-- x_5&#45;&gt;x_0 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>x_5&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M323.9,-71.7C328.27,-63.2 333.61,-52.81 338.39,-43.52\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"341.5,-45.12 342.96,-34.62 335.28,-41.92 341.5,-45.12\"/>\n",
       "</g>\n",
       "<!-- x_6 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>x_6</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"414,-108 360,-108 360,-72 414,-72 414,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"387\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;6</text>\n",
       "</g>\n",
       "<!-- x_6&#45;&gt;x_0 -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>x_6&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M378.1,-71.7C373.73,-63.2 368.39,-52.81 363.61,-43.52\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"366.72,-41.92 359.04,-34.62 360.5,-45.12 366.72,-41.92\"/>\n",
       "</g>\n",
       "<!-- x_7 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>x_7</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"486,-108 432,-108 432,-72 486,-72 486,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"459\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;7</text>\n",
       "</g>\n",
       "<!-- x_7&#45;&gt;x_0 -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>x_7&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M432.03,-71.52C415.37,-60.72 394.08,-46.92 377.51,-36.18\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"379.45,-33.27 369.15,-30.76 375.64,-39.14 379.45,-33.27\"/>\n",
       "</g>\n",
       "<!-- x_8 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>x_8</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"558,-108 504,-108 504,-72 558,-72 558,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"531\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;8</text>\n",
       "</g>\n",
       "<!-- x_8&#45;&gt;x_0 -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>x_8&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M503.6,-75.75C500.71,-74.45 497.8,-73.18 495,-72 458.29,-56.61 415.38,-41.13 386.01,-30.91\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"387.48,-27.72 376.89,-27.76 385.2,-34.33 387.48,-27.72\"/>\n",
       "</g>\n",
       "<!-- x_9 -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>x_9</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"630,-108 576,-108 576,-72 630,-72 630,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"603\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;9</text>\n",
       "</g>\n",
       "<!-- x_9&#45;&gt;x_0 -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>x_9&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M575.72,-75.43C572.81,-74.19 569.87,-73.02 567,-72 508.21,-51.14 437.59,-35.5 393.49,-26.8\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"394.3,-23.39 383.82,-24.92 392.97,-30.27 394.3,-23.39\"/>\n",
       "</g>\n",
       "<!-- x_10 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>x_10</title>\n",
       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"702,-108 648,-108 648,-72 702,-72 702,-108\"/>\n",
       "<text text-anchor=\"middle\" x=\"675\" y=\"-85.33\" font-family=\"Times,serif\" font-size=\"14.00\">Bi&#45;10</text>\n",
       "</g>\n",
       "<!-- x_10&#45;&gt;x_0 -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>x_10&#45;&gt;x_0</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M647.78,-75.25C644.86,-74.05 641.89,-72.93 639,-72 556.83,-45.54 456.96,-30.86 399.02,-23.98\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"399.5,-20.51 389.16,-22.84 398.7,-27.46 399.5,-20.51\"/>\n",
       "</g>\n",
       "<!-- x_11 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>x_11</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"53,-180 17,-180 17,-144 53,-144 53,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"35\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">11</text>\n",
       "</g>\n",
       "<!-- x_11&#45;&gt;x_1 -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>x_11&#45;&gt;x_1</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M33.02,-143.7C32.19,-136.41 31.2,-127.73 30.26,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"33.74,-119.15 29.13,-109.61 26.78,-119.94 33.74,-119.15\"/>\n",
       "</g>\n",
       "<!-- x_12 -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>x_12</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"120,-180 84,-180 84,-144 120,-144 120,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"102\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">12</text>\n",
       "</g>\n",
       "<!-- x_12&#45;&gt;x_2 -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>x_12&#45;&gt;x_2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M101.26,-143.7C100.95,-136.41 100.57,-127.73 100.22,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"103.72,-119.46 99.8,-109.62 96.73,-119.76 103.72,-119.46\"/>\n",
       "</g>\n",
       "<!-- x_13 -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>x_13</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"192,-180 156,-180 156,-144 192,-144 192,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"174\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">13</text>\n",
       "</g>\n",
       "<!-- x_13&#45;&gt;x_3 -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>x_13&#45;&gt;x_3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M173.26,-143.7C172.95,-136.41 172.57,-127.73 172.22,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"175.72,-119.46 171.8,-109.62 168.73,-119.76 175.72,-119.46\"/>\n",
       "</g>\n",
       "<!-- x_14 -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>x_14</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"263,-180 227,-180 227,-144 263,-144 263,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"245\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">14</text>\n",
       "</g>\n",
       "<!-- x_14&#45;&gt;x_4 -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>x_14&#45;&gt;x_4</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M244.51,-143.7C244.3,-136.41 244.05,-127.73 243.82,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"247.32,-119.51 243.53,-109.62 240.32,-119.71 247.32,-119.51\"/>\n",
       "</g>\n",
       "<!-- x_15 -->\n",
       "<g id=\"node16\" class=\"node\">\n",
       "<title>x_15</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"334,-180 298,-180 298,-144 334,-144 334,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"316\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">15</text>\n",
       "</g>\n",
       "<!-- x_15&#45;&gt;x_5 -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>x_15&#45;&gt;x_5</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M315.75,-143.7C315.65,-136.41 315.52,-127.73 315.41,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"318.91,-119.57 315.27,-109.62 311.91,-119.67 318.91,-119.57\"/>\n",
       "</g>\n",
       "<!-- x_16 -->\n",
       "<g id=\"node17\" class=\"node\">\n",
       "<title>x_16</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"404,-180 368,-180 368,-144 404,-144 404,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"386\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">16</text>\n",
       "</g>\n",
       "<!-- x_16&#45;&gt;x_6 -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>x_16&#45;&gt;x_6</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M386.25,-143.7C386.35,-136.41 386.48,-127.73 386.59,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"390.09,-119.67 386.73,-109.62 383.09,-119.57 390.09,-119.67\"/>\n",
       "</g>\n",
       "<!-- x_17 -->\n",
       "<g id=\"node18\" class=\"node\">\n",
       "<title>x_17</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"475,-180 439,-180 439,-144 475,-144 475,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"457\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">17</text>\n",
       "</g>\n",
       "<!-- x_17&#45;&gt;x_7 -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>x_17&#45;&gt;x_7</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M457.49,-143.7C457.7,-136.41 457.95,-127.73 458.18,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"461.68,-119.71 458.47,-109.62 454.68,-119.51 461.68,-119.71\"/>\n",
       "</g>\n",
       "<!-- x_18 -->\n",
       "<g id=\"node19\" class=\"node\">\n",
       "<title>x_18</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"547,-180 511,-180 511,-144 547,-144 547,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"529\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">18</text>\n",
       "</g>\n",
       "<!-- x_18&#45;&gt;x_8 -->\n",
       "<g id=\"edge18\" class=\"edge\">\n",
       "<title>x_18&#45;&gt;x_8</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M529.49,-143.7C529.7,-136.41 529.95,-127.73 530.18,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"533.68,-119.71 530.47,-109.62 526.68,-119.51 533.68,-119.71\"/>\n",
       "</g>\n",
       "<!-- x_19 -->\n",
       "<g id=\"node20\" class=\"node\">\n",
       "<title>x_19</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"619,-180 583,-180 583,-144 619,-144 619,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"601\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">19</text>\n",
       "</g>\n",
       "<!-- x_19&#45;&gt;x_9 -->\n",
       "<g id=\"edge19\" class=\"edge\">\n",
       "<title>x_19&#45;&gt;x_9</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M601.49,-143.7C601.7,-136.41 601.95,-127.73 602.18,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"605.68,-119.71 602.47,-109.62 598.68,-119.51 605.68,-119.71\"/>\n",
       "</g>\n",
       "<!-- x_20 -->\n",
       "<g id=\"node21\" class=\"node\">\n",
       "<title>x_20</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"691,-180 655,-180 655,-144 691,-144 691,-180\"/>\n",
       "<text text-anchor=\"middle\" x=\"673\" y=\"-157.32\" font-family=\"Times,serif\" font-size=\"14.00\">20</text>\n",
       "</g>\n",
       "<!-- x_20&#45;&gt;x_10 -->\n",
       "<g id=\"edge20\" class=\"edge\">\n",
       "<title>x_20&#45;&gt;x_10</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M673.49,-143.7C673.7,-136.41 673.95,-127.73 674.18,-119.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"677.68,-119.71 674.47,-109.62 670.68,-119.51 677.68,-119.71\"/>\n",
       "</g>\n",
       "<!-- x_21 -->\n",
       "<g id=\"node22\" class=\"node\">\n",
       "<title>x_21</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"38\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"38\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">21</text>\n",
       "</g>\n",
       "<!-- x_21&#45;&gt;x_11 -->\n",
       "<g id=\"edge21\" class=\"edge\">\n",
       "<title>x_21&#45;&gt;x_11</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M37.08,-215.68C36.79,-208.16 36.45,-199.72 36.14,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"39.64,-191.86 35.75,-182 32.65,-192.13 39.64,-191.86\"/>\n",
       "</g>\n",
       "<!-- x_22 -->\n",
       "<g id=\"node23\" class=\"node\">\n",
       "<title>x_22</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"105\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"105\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">22</text>\n",
       "</g>\n",
       "<!-- x_22&#45;&gt;x_12 -->\n",
       "<g id=\"edge22\" class=\"edge\">\n",
       "<title>x_22&#45;&gt;x_12</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M104.08,-215.68C103.79,-208.16 103.45,-199.72 103.14,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"106.64,-191.86 102.75,-182 99.65,-192.13 106.64,-191.86\"/>\n",
       "</g>\n",
       "<!-- x_23 -->\n",
       "<g id=\"node24\" class=\"node\">\n",
       "<title>x_23</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"177\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"177\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">23</text>\n",
       "</g>\n",
       "<!-- x_23&#45;&gt;x_13 -->\n",
       "<g id=\"edge23\" class=\"edge\">\n",
       "<title>x_23&#45;&gt;x_13</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M176.08,-215.68C175.79,-208.16 175.45,-199.72 175.14,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"178.64,-191.86 174.75,-182 171.65,-192.13 178.64,-191.86\"/>\n",
       "</g>\n",
       "<!-- x_24 -->\n",
       "<g id=\"node25\" class=\"node\">\n",
       "<title>x_24</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"248\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"248\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">24</text>\n",
       "</g>\n",
       "<!-- x_24&#45;&gt;x_14 -->\n",
       "<g id=\"edge24\" class=\"edge\">\n",
       "<title>x_24&#45;&gt;x_14</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M247.08,-215.68C246.79,-208.16 246.45,-199.72 246.14,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"249.64,-191.86 245.75,-182 242.65,-192.13 249.64,-191.86\"/>\n",
       "</g>\n",
       "<!-- x_25 -->\n",
       "<g id=\"node26\" class=\"node\">\n",
       "<title>x_25</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"317\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"317\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">25</text>\n",
       "</g>\n",
       "<!-- x_25&#45;&gt;x_15 -->\n",
       "<g id=\"edge25\" class=\"edge\">\n",
       "<title>x_25&#45;&gt;x_15</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M316.69,-215.68C316.6,-208.16 316.48,-199.72 316.38,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"319.88,-191.96 316.25,-182 312.88,-192.05 319.88,-191.96\"/>\n",
       "</g>\n",
       "<!-- x_26 -->\n",
       "<g id=\"node27\" class=\"node\">\n",
       "<title>x_26</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"386\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"386\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">26</text>\n",
       "</g>\n",
       "<!-- x_26&#45;&gt;x_16 -->\n",
       "<g id=\"edge26\" class=\"edge\">\n",
       "<title>x_26&#45;&gt;x_16</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M386,-215.68C386,-208.16 386,-199.72 386,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"389.5,-192 386,-182 382.5,-192 389.5,-192\"/>\n",
       "</g>\n",
       "<!-- x_27 -->\n",
       "<g id=\"node28\" class=\"node\">\n",
       "<title>x_27</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"456\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"456\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">27</text>\n",
       "</g>\n",
       "<!-- x_27&#45;&gt;x_17 -->\n",
       "<g id=\"edge27\" class=\"edge\">\n",
       "<title>x_27&#45;&gt;x_17</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M456.31,-215.68C456.4,-208.16 456.52,-199.72 456.62,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"460.12,-192.05 456.75,-182 453.12,-191.96 460.12,-192.05\"/>\n",
       "</g>\n",
       "<!-- x_28 -->\n",
       "<g id=\"node29\" class=\"node\">\n",
       "<title>x_28</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"527\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"527\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">28</text>\n",
       "</g>\n",
       "<!-- x_28&#45;&gt;x_18 -->\n",
       "<g id=\"edge28\" class=\"edge\">\n",
       "<title>x_28&#45;&gt;x_18</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M527.61,-215.68C527.81,-208.16 528.03,-199.72 528.24,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"531.73,-192.09 528.5,-182 524.74,-191.91 531.73,-192.09\"/>\n",
       "</g>\n",
       "<!-- x_29 -->\n",
       "<g id=\"node30\" class=\"node\">\n",
       "<title>x_29</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"599\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"599\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">29</text>\n",
       "</g>\n",
       "<!-- x_29&#45;&gt;x_19 -->\n",
       "<g id=\"edge29\" class=\"edge\">\n",
       "<title>x_29&#45;&gt;x_19</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M599.61,-215.68C599.81,-208.16 600.03,-199.72 600.24,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"603.73,-192.09 600.5,-182 596.74,-191.91 603.73,-192.09\"/>\n",
       "</g>\n",
       "<!-- x_30 -->\n",
       "<g id=\"node31\" class=\"node\">\n",
       "<title>x_30</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"671\" cy=\"-239.85\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"671\" y=\"-235.18\" font-family=\"Times,serif\" font-size=\"14.00\">30</text>\n",
       "</g>\n",
       "<!-- x_30&#45;&gt;x_20 -->\n",
       "<g id=\"edge30\" class=\"edge\">\n",
       "<title>x_30&#45;&gt;x_20</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M671.61,-215.68C671.81,-208.16 672.03,-199.72 672.24,-191.84\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"675.73,-192.09 672.5,-182 668.74,-191.91 675.73,-192.09\"/>\n",
       "</g>\n",
       "<!-- x_31 -->\n",
       "<g id=\"node32\" class=\"node\">\n",
       "<title>x_31</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"350\" cy=\"-323.55\" rx=\"23.85\" ry=\"23.85\"/>\n",
       "<text text-anchor=\"middle\" x=\"350\" y=\"-318.88\" font-family=\"Times,serif\" font-size=\"14.00\">31</text>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_21 -->\n",
       "<g id=\"edge31\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_21</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M326.02,-320.58C276.7,-315.86 161.05,-301.37 72,-263.7 69.38,-262.59 66.75,-261.27 64.17,-259.83\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"60.52\" cy=\"-257.6\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_22 -->\n",
       "<g id=\"edge32\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_22</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M326.48,-317.86C287.53,-309.64 207.61,-290.9 144,-263.7 140.14,-262.05 136.2,-260.1 132.37,-258.03\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"128.65\" cy=\"-255.93\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_23 -->\n",
       "<g id=\"edge33\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_23</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M328.09,-314.02C301.04,-303.27 253.86,-283.74 215,-263.7 211.52,-261.91 207.92,-259.92 204.39,-257.89\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"200.72\" cy=\"-255.73\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_24 -->\n",
       "<g id=\"edge34\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_24</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M331.76,-307.94C315.48,-294.9 291.45,-275.66 273.31,-261.12\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"269.93\" cy=\"-258.42\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_25 -->\n",
       "<g id=\"edge35\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_25</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M341.33,-301.09C337.57,-291.78 333.1,-280.72 329.03,-270.64\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"327.41\" cy=\"-266.63\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_26 -->\n",
       "<g id=\"edge36\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_26</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M359.27,-301.51C363.48,-291.96 368.53,-280.5 373.1,-270.13\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"374.82\" cy=\"-266.22\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_27 -->\n",
       "<g id=\"edge37\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_27</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M368.46,-308.32C385.61,-295.1 411.3,-275.3 430.42,-260.57\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"433.83\" cy=\"-257.94\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_28 -->\n",
       "<g id=\"edge38\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_28</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M372.1,-314.31C399.89,-303.69 448.83,-284.15 489,-263.7 492.49,-261.92 496.09,-259.95 499.63,-257.93\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"503.3\" cy=\"-255.77\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_29 -->\n",
       "<g id=\"edge39\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_29</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M373.66,-318C413.26,-309.92 495,-291.3 560,-263.7 563.86,-262.06 567.81,-260.11 571.64,-258.05\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"575.36\" cy=\"-255.95\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "<!-- x_31&#45;&gt;x_30 -->\n",
       "<g id=\"edge40\" class=\"edge\">\n",
       "<title>x_31&#45;&gt;x_30</title>\n",
       "<path fill=\"none\" stroke=\"gray\" stroke-dasharray=\"5,2\" d=\"M373.76,-320.22C423.23,-314.8 540.3,-299.04 632,-263.7 635.91,-262.19 639.89,-260.32 643.74,-258.3\"/>\n",
       "<ellipse fill=\"gray\" stroke=\"gray\" cx=\"647.46\" cy=\"-256.21\" rx=\"4\" ry=\"4\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.sources.Source at 0x7f7b33c3b5c0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "categorical_hgf.plot_network()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea060dd5-e679-4793-bfe2-78f3116f1b38",
   "metadata": {},
   "source": [
    "### Fitting the model forwards"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6e27b3d9-1b10-4cf9-9074-725a9638345d",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "categorical_hgf.input_data(input_data=input_data.T);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3e76f37e-36b0-49d4-8169-86f08b309e0d",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x900 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(nrows=5, figsize=(12, 9), sharex=True)\n",
    "plot_nodes(categorical_hgf, node_idxs=31, axs=axs[0])\n",
    "axs[1].imshow(\n",
    "    categorical_hgf.node_trajectories[0][\"mean\"].T, interpolation=\"none\", aspect=\"auto\"\n",
    ")\n",
    "axs[1].set_title(\"Mean of the implied Dirichlet distribution\", loc=\"left\")\n",
    "axs[1].set_ylabel(\"Categories\")\n",
    "\n",
    "# observations\n",
    "axs[2].imshow(input_data, interpolation=\"none\", aspect=\"auto\", cmap=\"binary\")\n",
    "axs[2].set_title(\"Observations\", loc=\"left\")\n",
    "axs[2].set_ylabel(\"Categories\")\n",
    "\n",
    "# surprise\n",
    "axs[3].plot(\n",
    "    np.arange(2, len(categorical_hgf.node_trajectories[0][\"kl_divergence\"])),\n",
    "    categorical_hgf.node_trajectories[0][\"kl_divergence\"][2:],\n",
    "    color=\"#2a2a2a\",\n",
    "    linewidth=0.5,\n",
    "    zorder=-1,\n",
    "    label=\"Surprise\",\n",
    ")\n",
    "axs[3].fill_between(\n",
    "    x=np.arange(2, len(categorical_hgf.node_trajectories[0][\"kl_divergence\"])),\n",
    "    y1=categorical_hgf.node_trajectories[0][\"kl_divergence\"][2:],\n",
    "    y2=categorical_hgf.node_trajectories[0][\"kl_divergence\"][2:].min(),\n",
    "    color=\"#7f7f7f\",\n",
    "    alpha=0.1,\n",
    "    zorder=-1,\n",
    ")\n",
    "axs[3].set_title(\"Kullback-Leibler divergences\", loc=\"left\")\n",
    "axs[2].set_ylabel(\"Surprises\")\n",
    "\n",
    "axs[4].plot(\n",
    "    np.arange(len(categorical_hgf.node_trajectories[0][\"surprise\"])),\n",
    "    categorical_hgf.node_trajectories[0][\"surprise\"],\n",
    "    color=\"#2a2a2a\",\n",
    "    linewidth=0.5,\n",
    "    zorder=-1,\n",
    "    label=\"Surprise\",\n",
    ")\n",
    "axs[4].fill_between(\n",
    "    x=np.arange(len(categorical_hgf.node_trajectories[0][\"surprise\"])),\n",
    "    y1=categorical_hgf.node_trajectories[0][\"surprise\"],\n",
    "    y2=categorical_hgf.node_trajectories[0][\"surprise\"].min(),\n",
    "    color=\"#7f7f7f\",\n",
    "    alpha=0.1,\n",
    "    zorder=-1,\n",
    ")\n",
    "axs[4].set_title(\"Sum of binary surprises\", loc=\"left\")\n",
    "axs[2].set_ylabel(\"Surprises\")\n",
    "\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d915da54-5396-467c-827b-583a77ff8391",
   "metadata": {},
   "source": [
    "### Inference using MCMC sampling"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aea0e621-26f2-436e-b958-f0e9caa50992",
   "metadata": {},
   "source": [
    "In the binary and continuous HGF example, we have been using the {py:class}`pyhgf.distribution.HGFDistribution` class to create a PyMC-compatible distribution of the HGF. This was possible when using the most standard models as we can easily write a pre-defined distribution that fits exactly the network specification. However, when using more exotic network structures, as this is the case here with the categorical state nodes where the number of nodes in the network grows with the number of categories, we need a more flexible approach that can let us wrap a PyMC distribution for every kind of network we can have. \n",
    "\n",
    "This is what we are doing below (see [this blog post](https://www.pymc-labs.io/blog-posts/jax-functions-in-pymc-3-quick-examples/) and the [PyMC documentation](https://www.pymc.io/projects/examples/en/latest/case_studies/wrapping_jax_function.html) on how to do that). First, we start by creating a function that computes the surprise of the model, here using the Kullback-Leibler divergences of the implied Dirichlet distributions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "895b59fc-9e49-4848-bc07-59325e34b5d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def categorical_surprise(omega_2, hgf, input_data):\n",
    "\n",
    "    # replace with a new omega in the model\n",
    "    for idx in np.arange(21, 31):\n",
    "        hgf.attributes[idx][\"tonic_volatility\"] = omega_2\n",
    "\n",
    "    # run the model forward again\n",
    "    hgf.input_data(input_data=input_data.T)\n",
    "\n",
    "    # compute the surprises using KL divergences\n",
    "    surprise = hgf.node_trajectories[0][\"kl_divergence\"][2:].sum()\n",
    "\n",
    "    # return an infinite surprise if the model could not fit at any point\n",
    "    surprise = jnp.where(\n",
    "        jnp.any(jnp.isnan(hgf.node_trajectories[0][\"xi\"])), jnp.inf, surprise\n",
    "    )\n",
    "\n",
    "    return surprise\n",
    "\n",
    "\n",
    "surprise_fn = Partial(categorical_surprise, hgf=categorical_hgf, input_data=input_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5ebba47-0e0e-47ca-9c8f-91dc84422394",
   "metadata": {},
   "source": [
    "We create both jitted and the vector-jacobian product requiered for a custom Op in PyTensor:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "05bcec35-1b57-4593-9285-b773f0f9e5c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "jitted_custom_op_jax = jit(surprise_fn)\n",
    "\n",
    "\n",
    "def vjp_custom_op_jax(x, gz):\n",
    "    _, vjp_fn = vjp(surprise_fn, x)\n",
    "    return vjp_fn(gz)[0]\n",
    "\n",
    "\n",
    "jitted_vjp_custom_op_jax = jit(vjp_custom_op_jax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "191f65ab-6b82-4de6-9a3a-940f5d6c7400",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": [
     "hide-cell"
    ]
   },
   "outputs": [],
   "source": [
    "# The CustomOp needs `make_node`, `perform` and `grad`.\n",
    "class CustomOp(Op):\n",
    "    def make_node(self, x):\n",
    "        # Create a PyTensor node specifying the number and type of inputs and outputs\n",
    "\n",
    "        # We convert the input into a PyTensor tensor variable\n",
    "        inputs = [pt.as_tensor_variable(x)]\n",
    "        # Output has the same type and shape as `x`\n",
    "        outputs = [inputs[0].type()]\n",
    "        return Apply(self, inputs, outputs)\n",
    "\n",
    "    def perform(self, node, inputs, outputs):\n",
    "        # Evaluate the Op result for a specific numerical input\n",
    "\n",
    "        # The inputs are always wrapped in a list\n",
    "        (x,) = inputs\n",
    "        result = jitted_custom_op_jax(x)\n",
    "        # The results should be assigned inplace to the nested list\n",
    "        # of outputs provided by PyTensor. If you have multiple\n",
    "        # outputs and results, you should assign each at outputs[i][0]\n",
    "        outputs[0][0] = np.asarray(result, dtype=\"float64\")\n",
    "\n",
    "    def grad(self, inputs, output_gradients):\n",
    "        # Create a PyTensor expression of the gradient\n",
    "        (x,) = inputs\n",
    "        (gz,) = output_gradients\n",
    "        # We reference the VJP Op created below, which encapsulates\n",
    "        # the gradient operation\n",
    "        return [vjp_custom_op(x, gz)]\n",
    "\n",
    "\n",
    "class VJPCustomOp(Op):\n",
    "    def make_node(self, x, gz):\n",
    "        # Make sure the two inputs are tensor variables\n",
    "        inputs = [pt.as_tensor_variable(x), pt.as_tensor_variable(gz)]\n",
    "        # Output has the shape type and shape as the first input\n",
    "        outputs = [inputs[0].type()]\n",
    "        return Apply(self, inputs, outputs)\n",
    "\n",
    "    def perform(self, node, inputs, outputs):\n",
    "        (x, gz) = inputs\n",
    "        result = jitted_vjp_custom_op_jax(x, gz)\n",
    "        outputs[0][0] = np.asarray(result, dtype=\"float64\")\n",
    "\n",
    "\n",
    "# Instantiate the Ops\n",
    "custom_op = CustomOp()\n",
    "vjp_custom_op = VJPCustomOp()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ccf51208-980b-4ccd-9b2a-cae1525da3f0",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# with pm.Model() as model:\n",
    "#    omega_2 = pm.Normal(\"omega_2\", -2.0, 2)\n",
    "#    pm.Potential(\"hgf\", custom_op(omega_2))\n",
    "#    categorical_idata = pm.sample(chains=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bce093e9-1ab0-4ea1-99e6-a83b68331b9c",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# az.plot_trace(categorical_idata)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44abf936-9aba-487c-bbb0-96ed9af3de73",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "source": [
    "## The categorical state-transition node\n",
    "\n",
    "```{warning}\n",
    "This is work in progress.\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3c8e69f-b9f6-4c42-ac1e-f31bfb6b1bf3",
   "metadata": {},
   "source": [
    "# System configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "25f8522f-9d42-47fd-a6bc-87ffc0dfd866",
   "metadata": {
    "editable": true,
    "scrolled": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Last updated: Fri Aug 23 2024\n",
      "\n",
      "Python implementation: CPython\n",
      "Python version       : 3.12.3\n",
      "IPython version      : 8.23.0\n",
      "\n",
      "pyhgf : 0.1.4\n",
      "jax   : 0.4.31\n",
      "jaxlib: 0.4.31\n",
      "\n",
      "pytensor  : 2.25.2\n",
      "matplotlib: 3.8.4\n",
      "jax       : 0.4.31\n",
      "numpy     : 1.26.0\n",
      "sys       : 3.12.3 | packaged by conda-forge | (main, Apr 15 2024, 18:38:13) [GCC 12.3.0]\n",
      "seaborn   : 0.13.2\n",
      "\n",
      "Watermark: 2.4.3\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -n -u -v -iv -w -p pyhgf,jax,jaxlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f5351f55-8c0a-4e45-b5d5-32853b4b3e2e",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}