{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Attempt at an ordered probit model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import pprint\n",
    "import logging\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import pymc3 as pm\n",
    "import pymc3.theanof\n",
    "\n",
    "import theano.tensor as tt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.8\n"
     ]
    }
   ],
   "source": [
    "print(pm.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define an `OrderedProbit` distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# as per pm.OrderedLogistic\n",
    "class OrderedProbit(pm.Categorical):\n",
    "    def __init__(self, mu, sigma, cutpoints, *args, **kwargs):\n",
    "\n",
    "        mu = tt.as_tensor_variable(pymc3.theanof.floatX(mu))\n",
    "        sigma = tt.as_tensor_variable(pymc3.theanof.floatX(sigma))\n",
    "        cutpoints = tt.as_tensor_variable(cutpoints)\n",
    "\n",
    "        p = (\n",
    "            self.phi(x=cutpoints[:, 1:], mu=mu, sigma=sigma) -\n",
    "            self.phi(x=cutpoints[:, :-1], mu=mu, sigma=sigma)\n",
    "        )\n",
    "\n",
    "        # must be >= 0\n",
    "        p = tt.where(p <= 0, 1e-50, p)\n",
    "\n",
    "        super().__init__(p=p, *args, **kwargs)\n",
    "\n",
    "    @staticmethod\n",
    "    def phi(x, mu, sigma):\n",
    "        p = 0.5 * (1.0 + pm.math.erf(((x - mu) / sigma) / pm.math.sqrt(2)))\n",
    "        return p"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define the response scale and cutpoints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "array([[-inf,  1.5,  2.5,  3.5,  4.5,  5.5,  6.5,  7.5,  8.5,  9.5,  inf]])\n"
     ]
    }
   ],
   "source": [
    "# can choose 1 to 10\n",
    "response_options = np.arange(1, 11)\n",
    "\n",
    "# evenly spaced cutpoints\n",
    "cutpoints = np.concatenate(([-np.Inf], (response_options - 0.5)[1:], [+np.Inf]))[np.newaxis, :]\n",
    "pprint.pprint(cutpoints)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_trials = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_model(y=None):\n",
    "    \n",
    "    with pm.Model() as model:\n",
    "        \n",
    "        mu = pm.Normal(\"mu\", mu=np.mean(response_options), sigma=2.0)\n",
    "        sigma = pm.Lognormal(\"sigma\", mu=0.5, sigma=0.5)\n",
    "        \n",
    "        OrderedProbit(\"y\", mu=mu, sigma=sigma, cutpoints=cutpoints, observed=y, shape=n_trials)\n",
    "        \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = get_model()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Draw a sample from the prior"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "with model:\n",
    "    draw = pm.sample_prior_predictive(samples=1, random_seed=1234)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'mu': array(6.44287033),\n",
      " 'sigma': array(0.90892941),\n",
      " 'sigma_log__': -0.0954878473532322,\n",
      " 'y': array([5, 6, 6, 5, 5, 6, 7, 6, 5, 5, 6, 6, 5, 6, 5, 3, 6, 7, 5, 6, 4, 5,\n",
      "       7, 6, 5, 6, 5, 6, 6, 5, 6, 4, 6, 6, 5, 7, 5, 7, 4, 5, 4, 6, 6, 6,\n",
      "       4, 6, 5, 5, 4, 6])}\n"
     ]
    }
   ],
   "source": [
    "pprint.pprint(draw)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(draw[\"y\"] + 1);\n",
    "plt.xlim([response_options[0], response_options[-1]]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define the model with the prior draw as observed data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "obs_model = get_model(y=draw[\"y\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Try sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Sequential sampling (1 chains in 1 job)\n",
      "NUTS: [sigma, mu]\n",
      "Sampling chain 0, 0 divergences:  10%|█         | 103/1000 [00:02<00:17, 50.59it/s]\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "Mass matrix contains zeros on the diagonal. \nThe derivative of RV `sigma_log__`.ravel()[0] is zero.\nThe derivative of RV `mu`.ravel()[0] is zero.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-12-8b6cf7bec66f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mobs_model\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m     \u001b[0mtrace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msample\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcores\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchains\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/sampling.py\u001b[0m in \u001b[0;36msample\u001b[0;34m(draws, step, init, n_init, start, trace, chain_idx, chains, cores, tune, progressbar, model, random_seed, discard_tuned_samples, compute_convergence_checks, **kwargs)\u001b[0m\n\u001b[1;32m    487\u001b[0m             \u001b[0m_log\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minfo\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Sequential sampling ({} chains in 1 job)\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchains\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    488\u001b[0m             \u001b[0m_print_step_hierarchy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 489\u001b[0;31m             \u001b[0mtrace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_sample_many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0msample_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    490\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    491\u001b[0m     \u001b[0mdiscard\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtune\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mdiscard_tuned_samples\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/sampling.py\u001b[0m in \u001b[0;36m_sample_many\u001b[0;34m(draws, chain, chains, start, random_seed, step, **kwargs)\u001b[0m\n\u001b[1;32m    531\u001b[0m     \u001b[0mtraces\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    532\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchains\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 533\u001b[0;31m         trace = _sample(\n\u001b[0m\u001b[1;32m    534\u001b[0m             \u001b[0mdraws\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdraws\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    535\u001b[0m             \u001b[0mchain\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mchain\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/sampling.py\u001b[0m in \u001b[0;36m_sample\u001b[0;34m(chain, progressbar, random_seed, start, draws, step, trace, tune, model, **kwargs)\u001b[0m\n\u001b[1;32m    603\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    604\u001b[0m         \u001b[0mstrace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 605\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mstrace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdiverging\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msampling\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    606\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mit\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mskip_first\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    607\u001b[0m                 \u001b[0mtrace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mMultiTrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstrace\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/tqdm/std.py\u001b[0m in \u001b[0;36m__iter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1125\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1126\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1127\u001b[0;31m             \u001b[0;32mfor\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miterable\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1128\u001b[0m                 \u001b[0;32myield\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1129\u001b[0m                 \u001b[0;31m# Update and possibly print the progressbar.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/sampling.py\u001b[0m in \u001b[0;36m_iter_sample\u001b[0;34m(draws, step, start, trace, chain, tune, model, random_seed)\u001b[0m\n\u001b[1;32m    698\u001b[0m                 \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstop_tuning\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    699\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgenerates_stats\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 700\u001b[0;31m                 \u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstats\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpoint\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    701\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mstrace\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msupports_sampler_stats\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    702\u001b[0m                     \u001b[0mstrace\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecord\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstats\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/step_methods/arraystep.py\u001b[0m in \u001b[0;36mstep\u001b[0;34m(self, point)\u001b[0m\n\u001b[1;32m    245\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    246\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgenerates_stats\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 247\u001b[0;31m             \u001b[0mapoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstats\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    248\u001b[0m             \u001b[0mpoint\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_logp_dlogp_func\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray_to_full_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mapoint\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    249\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstats\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/step_methods/hmc/base_hmc.py\u001b[0m in \u001b[0;36mastep\u001b[0;34m(self, q0)\u001b[0m\n\u001b[1;32m    128\u001b[0m                 \u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mabs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcheck_test_point\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0;36m1e20\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m|\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misnan\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcheck_test_point\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    129\u001b[0m             ]\n\u001b[0;32m--> 130\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpotential\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraise_ok\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_logp_dlogp_func\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_ordering\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvmap\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    131\u001b[0m             message_energy = (\n\u001b[1;32m    132\u001b[0m                 \u001b[0;34m\"Bad initial energy, check any log probabilities that \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venv/sound_field_vr_ana/lib/python3.8/site-packages/pymc3/step_methods/hmc/quadpotential.py\u001b[0m in \u001b[0;36mraise_ok\u001b[0;34m(self, vmap)\u001b[0m\n\u001b[1;32m    229\u001b[0m                 errmsg.append('The derivative of RV `{}`.ravel()[{}]'\n\u001b[1;32m    230\u001b[0m                               ' is zero.'.format(*name_slc[ii]))\n\u001b[0;32m--> 231\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'\\n'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0merrmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    232\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    233\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m~\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misfinite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Mass matrix contains zeros on the diagonal. \nThe derivative of RV `sigma_log__`.ravel()[0] is zero.\nThe derivative of RV `mu`.ravel()[0] is zero."
     ]
    }
   ],
   "source": [
    "with obs_model:\n",
    "    trace = pm.sample(cores=1, chains=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Try removing `np.inf`s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "array([[-1.0e+51,  1.5e+00,  2.5e+00,  3.5e+00,  4.5e+00,  5.5e+00,\n",
      "         6.5e+00,  7.5e+00,  8.5e+00,  9.5e+00,  1.0e+51]])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sequential sampling (1 chains in 1 job)\n",
      "NUTS: [sigma, mu]\n",
      "Sampling chain 0, 0 divergences: 100%|██████████| 1000/1000 [00:00<00:00, 1997.36it/s]\n",
      "Only one chain was sampled, this makes it impossible to run some convergence checks\n"
     ]
    }
   ],
   "source": [
    "# evenly spaced cutpoints\n",
    "cutpoints = np.concatenate(([-10e50], (response_options - 0.5)[1:], [+10e50]))[np.newaxis, :]\n",
    "pprint.pprint(cutpoints)\n",
    "\n",
    "obs_model = get_model(y=draw[\"y\"])\n",
    "\n",
    "with obs_model:\n",
    "    trace = pm.sample(cores=1, chains=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 500), minimum_shape: (chains=2, draws=4)\n",
      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 500), minimum_shape: (chains=2, draws=4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hpd_3%</th>\n",
       "      <th>hpd_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_mean</th>\n",
       "      <th>ess_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mu</th>\n",
       "      <td>6.439</td>\n",
       "      <td>0.128</td>\n",
       "      <td>6.223</td>\n",
       "      <td>6.682</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.005</td>\n",
       "      <td>372.0</td>\n",
       "      <td>372.0</td>\n",
       "      <td>380.0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.899</td>\n",
       "      <td>0.105</td>\n",
       "      <td>0.723</td>\n",
       "      <td>1.100</td>\n",
       "      <td>0.006</td>\n",
       "      <td>0.004</td>\n",
       "      <td>334.0</td>\n",
       "      <td>334.0</td>\n",
       "      <td>292.0</td>\n",
       "      <td>320.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hpd_3%  hpd_97%  mcse_mean  mcse_sd  ess_mean  ess_sd  \\\n",
       "mu     6.439  0.128   6.223    6.682      0.007    0.005     372.0   372.0   \n",
       "sigma  0.899  0.105   0.723    1.100      0.006    0.004     334.0   334.0   \n",
       "\n",
       "       ess_bulk  ess_tail  r_hat  \n",
       "mu        380.0     244.0    NaN  \n",
       "sigma     292.0     320.0    NaN  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.summary(trace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.traceplot(trace);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Try sampling using `pm.sample_smc()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sample initial stage: ...\n",
      "Stage:   0 Beta: 0.040 Steps:  25 Acce: 1.000\n",
      "Stage:   1 Beta: 0.129 Steps:  25 Acce: 0.531\n",
      "Stage:   2 Beta: 0.363 Steps:   6 Acce: 0.440\n",
      "Stage:   3 Beta: 1.000 Steps:   7 Acce: 0.359\n"
     ]
    }
   ],
   "source": [
    "with obs_model:\n",
    "    trace = pm.sample_smc()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 1000), minimum_shape: (chains=2, draws=4)\n",
      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 1000), minimum_shape: (chains=2, draws=4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hpd_3%</th>\n",
       "      <th>hpd_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_mean</th>\n",
       "      <th>ess_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mu</th>\n",
       "      <td>6.461</td>\n",
       "      <td>0.135</td>\n",
       "      <td>6.223</td>\n",
       "      <td>6.721</td>\n",
       "      <td>0.004</td>\n",
       "      <td>0.003</td>\n",
       "      <td>914.0</td>\n",
       "      <td>914.0</td>\n",
       "      <td>918.0</td>\n",
       "      <td>828.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.901</td>\n",
       "      <td>0.101</td>\n",
       "      <td>0.727</td>\n",
       "      <td>1.092</td>\n",
       "      <td>0.003</td>\n",
       "      <td>0.002</td>\n",
       "      <td>979.0</td>\n",
       "      <td>979.0</td>\n",
       "      <td>913.0</td>\n",
       "      <td>932.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hpd_3%  hpd_97%  mcse_mean  mcse_sd  ess_mean  ess_sd  \\\n",
       "mu     6.461  0.135   6.223    6.721      0.004    0.003     914.0   914.0   \n",
       "sigma  0.901  0.101   0.727    1.092      0.003    0.002     979.0   979.0   \n",
       "\n",
       "       ess_bulk  ess_tail  r_hat  \n",
       "mu        918.0     828.0    NaN  \n",
       "sigma     913.0     932.0    NaN  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.summary(trace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.traceplot(trace);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `pm.sample_smc()` a few times to check"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run it a few times to check it wasn't just lucky\n",
    "n_reps = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "with model:\n",
    "    draws = pm.sample_prior_predictive(samples=n_reps, random_seed=12345)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# temporarily turn off the output\n",
    "logger = logging.getLogger(\"pymc3\")\n",
    "start_level = logger.level\n",
    "logger.setLevel(logging.CRITICAL)\n",
    "\n",
    "est_mus = np.full(n_reps, np.nan)\n",
    "est_sigmas = np.full(n_reps, np.nan)\n",
    "\n",
    "for i_rep in range(n_reps):\n",
    "    \n",
    "    obs_model = get_model(y=draws[\"y\"][i_rep, :])\n",
    "    \n",
    "    with obs_model:\n",
    "        trace = pm.sample_smc()\n",
    "        \n",
    "    est_mus[i_rep] = np.mean(trace[\"mu\"])\n",
    "    est_sigmas[i_rep] = np.mean(trace[\"sigma\"])\n",
    "    \n",
    "logger.setLevel(start_level)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(draws[\"mu\"], est_mus);\n",
    "plt.xlim([0, 10]);\n",
    "plt.ylim([0, 10]);\n",
    "plt.plot([0, 10], [0, 10]);\n",
    "plt.xlabel(\"True $\\mu$\");\n",
    "plt.ylabel(\"Estimated $\\mu$\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(draws[\"sigma\"], est_sigmas);\n",
    "plt.xlim([0, 10]);\n",
    "plt.ylim([0, 10]);\n",
    "plt.plot([0, 10], [0, 10]);\n",
    "plt.xlabel(\"True $\\sigma$\");\n",
    "plt.ylabel(\"Estimated $\\sigma$\");"
   ]
  }
 ],
 "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": "3.8.2"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "292.8px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}