Lecture25_03 - Mixtures and MCMC.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Mixtures and MCMC\n",
    "\n",
    "##### Keywords: supervised learning, semi-supervised learning, unsupervised learning, mixture model, gaussian mixture model, pymc3, label-switching, identifiability, normal distribution, pymc3 potentials\n",
    "\n",
    "We now do a study of learning mixture models with MCMC. We have already done this in the case of the Zero-Inflated Poisson Model, and will stick to Gaussian Mixture models for now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import scipy as sp\n",
    "import matplotlib as mpl\n",
    "import matplotlib.cm as cm\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "pd.set_option('display.width', 500)\n",
    "pd.set_option('display.max_columns', 100)\n",
    "pd.set_option('display.notebook_repr_html', True)\n",
    "import seaborn as sns\n",
    "sns.set_style(\"whitegrid\")\n",
    "sns.set_context(\"poster\")\n",
    "import pymc3 as pm\n",
    "import theano.tensor as tt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mixture of 2 Gaussians, the old faithful data\n",
    "\n",
    "We start by considering waiting times from the Old-Faithful Geyser at Yellowstone National Park."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "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>eruptions</th>\n",
       "      <th>waiting</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.600</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.800</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.333</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.283</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.533</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   eruptions  waiting\n",
       "0      3.600       79\n",
       "1      1.800       54\n",
       "2      3.333       74\n",
       "3      2.283       62\n",
       "4      4.533       85"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ofdata=pd.read_csv(\"data/oldfaithful.csv\")\n",
    "ofdata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n",
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(ofdata.waiting);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visually, there seem to be two components to the waiting time, so let us model this using a mixture of two gaussians. Remember that this is a unsupervized model, and all we are doing is modelling $p(x)$ , with the assumption that there are two clusters and a hidden variable $z$ that indexes them.\n",
    "\n",
    "Notice that these gaussians seem well separated. The separation of gaussians impacts how your sampler will perform."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "with pm.Model() as ofmodel:\n",
    "    p1 = pm.Uniform('p', 0, 1)\n",
    "    p2 = 1 - p1\n",
    "    p = tt.stack([p1, p2])\n",
    "    assignment = pm.Categorical(\"assignment\", p, \n",
    "                                shape=ofdata.shape[0])\n",
    "    sds = pm.Uniform(\"sds\", 0, 40, shape=2)\n",
    "    centers = pm.Normal(\"centers\", \n",
    "                        mu=np.array([50, 80]), \n",
    "                        sd=np.array([20, 20]), \n",
    "                        shape=2)\n",
    "    \n",
    "    # and to combine it with the observations:\n",
    "    observations = pm.Normal(\"obs\", mu=centers[assignment], sd=sds[assignment], observed=ofdata.waiting)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n",
      "CompoundStep\n",
      ">NUTS: [centers, sds, p]\n",
      ">BinaryGibbsMetropolis: [assignment]\n",
      "Sampling 2 chains: 100%|██████████| 21000/21000 [04:43<00:00, 73.96draws/s]\n"
     ]
    }
   ],
   "source": [
    "with ofmodel:\n",
    "    #step1 = pm.Metropolis(vars=[p, sds, centers])\n",
    "    #step2 = pm.CategoricalGibbsMetropolis(vars=[assignment])\n",
    "    #oftrace_full = pm.sample(10000, step=[step1, step2])\n",
    "    oftrace_full = pm.sample(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "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 2.38.0 (20140413.2041)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"546pt\" height=\"228pt\"\n",
       " viewBox=\"0.00 0.00 546.00 228.00\" 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(4 224)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-224 542,-224 542,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\"><title>cluster272</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M20,-8C20,-8 212,-8 212,-8 218,-8 224,-14 224,-20 224,-20 224,-144 224,-144 224,-150 218,-156 212,-156 212,-156 20,-156 20,-156 14,-156 8,-150 8,-144 8,-144 8,-20 8,-20 8,-14 14,-8 20,-8\"/>\n",
       "<text text-anchor=\"middle\" x=\"205.5\" y=\"-14.8\" font-family=\"Times,serif\" font-size=\"14.00\">272</text>\n",
       "</g>\n",
       "<g id=\"clust2\" class=\"cluster\"><title>cluster2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M244,-82C244,-82 518,-82 518,-82 524,-82 530,-88 530,-94 530,-94 530,-144 530,-144 530,-150 524,-156 518,-156 518,-156 244,-156 244,-156 238,-156 232,-150 232,-144 232,-144 232,-94 232,-94 232,-88 238,-82 244,-82\"/>\n",
       "<text text-anchor=\"middle\" x=\"518.5\" y=\"-88.8\" font-family=\"Times,serif\" font-size=\"14.00\">2</text>\n",
       "</g>\n",
       "<!-- p -->\n",
       "<g id=\"node1\" class=\"node\"><title>p</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"116\" cy=\"-202\" rx=\"54.3812\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"116\" y=\"-197.8\" font-family=\"Times,serif\" font-size=\"14.00\">p ~ Uniform</text>\n",
       "</g>\n",
       "<!-- assignment -->\n",
       "<g id=\"node2\" class=\"node\"><title>assignment</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"116\" cy=\"-130\" rx=\"99.7258\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"116\" y=\"-125.8\" font-family=\"Times,serif\" font-size=\"14.00\">assignment ~ Categorical</text>\n",
       "</g>\n",
       "<!-- p&#45;&gt;assignment -->\n",
       "<g id=\"edge1\" class=\"edge\"><title>p&#45;&gt;assignment</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M116,-183.697C116,-175.983 116,-166.712 116,-158.112\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"119.5,-158.104 116,-148.104 112.5,-158.104 119.5,-158.104\"/>\n",
       "</g>\n",
       "<!-- obs -->\n",
       "<g id=\"node3\" class=\"node\"><title>obs</title>\n",
       "<ellipse fill=\"lightgrey\" stroke=\"black\" cx=\"157\" cy=\"-56\" rx=\"58.7243\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"157\" y=\"-51.8\" font-family=\"Times,serif\" font-size=\"14.00\">obs ~ Normal</text>\n",
       "</g>\n",
       "<!-- assignment&#45;&gt;obs -->\n",
       "<g id=\"edge2\" class=\"edge\"><title>assignment&#45;&gt;obs</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M125.716,-111.937C130.669,-103.239 136.796,-92.4799 142.311,-82.7948\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"145.446,-84.3623 147.353,-73.9405 139.363,-80.8984 145.446,-84.3623\"/>\n",
       "</g>\n",
       "<!-- sds -->\n",
       "<g id=\"node4\" class=\"node\"><title>sds</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"461\" cy=\"-130\" rx=\"61.1468\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"461\" y=\"-125.8\" font-family=\"Times,serif\" font-size=\"14.00\">sds ~ Uniform</text>\n",
       "</g>\n",
       "<!-- sds&#45;&gt;obs -->\n",
       "<g id=\"edge3\" class=\"edge\"><title>sds&#45;&gt;obs</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M442.193,-112.65C428.842,-102.023 409.984,-88.9051 391,-82 361.39,-71.2296 282.517,-64.3624 224.491,-60.6047\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"224.481,-57.0972 214.28,-59.9614 224.04,-64.0834 224.481,-57.0972\"/>\n",
       "</g>\n",
       "<!-- centers -->\n",
       "<g id=\"node5\" class=\"node\"><title>centers</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"311\" cy=\"-130\" rx=\"71.2356\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"311\" y=\"-125.8\" font-family=\"Times,serif\" font-size=\"14.00\">centers ~ Normal</text>\n",
       "</g>\n",
       "<!-- centers&#45;&gt;obs -->\n",
       "<g id=\"edge4\" class=\"edge\"><title>centers&#45;&gt;obs</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M285.449,-112.972C269.342,-103.264 247.897,-91.0506 228,-82 220.891,-78.7661 213.231,-75.6796 205.666,-72.848\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"206.776,-69.5275 196.182,-69.4055 204.387,-76.1075 206.776,-69.5275\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.dot.Digraph at 0x1360947f0>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.model_to_graphviz(ofmodel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_base.py:3604: MatplotlibDeprecationWarning: \n",
      "The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
      "  alternative='`bottom`', obj_type='argument')\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "oftrace = oftrace_full[2000::5]\n",
    "pm.traceplot(oftrace);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "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>mc_error</th>\n",
       "      <th>hpd_2.5</th>\n",
       "      <th>hpd_97.5</th>\n",
       "      <th>n_eff</th>\n",
       "      <th>Rhat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>assignment__0</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__2</th>\n",
       "      <td>0.993125</td>\n",
       "      <td>0.082630</td>\n",
       "      <td>0.001504</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3248.753445</td>\n",
       "      <td>0.999745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__3</th>\n",
       "      <td>0.050937</td>\n",
       "      <td>0.219870</td>\n",
       "      <td>0.004231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3173.388070</td>\n",
       "      <td>1.000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__4</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__5</th>\n",
       "      <td>0.000313</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__6</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__7</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__8</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__9</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__10</th>\n",
       "      <td>0.000625</td>\n",
       "      <td>0.024992</td>\n",
       "      <td>0.000437</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3208.032629</td>\n",
       "      <td>0.999687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__11</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__12</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__13</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__14</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__15</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__16</th>\n",
       "      <td>0.046875</td>\n",
       "      <td>0.211371</td>\n",
       "      <td>0.003711</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3205.553814</td>\n",
       "      <td>0.999696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__17</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__18</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__19</th>\n",
       "      <td>0.999375</td>\n",
       "      <td>0.024992</td>\n",
       "      <td>0.000437</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3208.032629</td>\n",
       "      <td>0.999687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__20</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__21</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__22</th>\n",
       "      <td>0.999062</td>\n",
       "      <td>0.030604</td>\n",
       "      <td>0.000533</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3209.379383</td>\n",
       "      <td>0.999792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__23</th>\n",
       "      <td>0.817500</td>\n",
       "      <td>0.386256</td>\n",
       "      <td>0.006151</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3120.023664</td>\n",
       "      <td>0.999729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__24</th>\n",
       "      <td>0.993125</td>\n",
       "      <td>0.082630</td>\n",
       "      <td>0.001629</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3235.543269</td>\n",
       "      <td>1.001749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__25</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__26</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__27</th>\n",
       "      <td>0.997188</td>\n",
       "      <td>0.052958</td>\n",
       "      <td>0.001177</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3222.014229</td>\n",
       "      <td>0.999722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__28</th>\n",
       "      <td>0.998437</td>\n",
       "      <td>0.039498</td>\n",
       "      <td>0.000681</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3213.694327</td>\n",
       "      <td>0.999750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__29</th>\n",
       "      <td>0.998750</td>\n",
       "      <td>0.035333</td>\n",
       "      <td>0.000612</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3212.072926</td>\n",
       "      <td>0.999687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__247</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__248</th>\n",
       "      <td>0.576562</td>\n",
       "      <td>0.494103</td>\n",
       "      <td>0.008319</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2947.001300</td>\n",
       "      <td>1.000571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__249</th>\n",
       "      <td>0.995938</td>\n",
       "      <td>0.063608</td>\n",
       "      <td>0.001140</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2794.058393</td>\n",
       "      <td>0.999712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__250</th>\n",
       "      <td>0.000625</td>\n",
       "      <td>0.024992</td>\n",
       "      <td>0.000437</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__251</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__252</th>\n",
       "      <td>0.983750</td>\n",
       "      <td>0.126436</td>\n",
       "      <td>0.002048</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3312.296435</td>\n",
       "      <td>0.999907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__253</th>\n",
       "      <td>0.983125</td>\n",
       "      <td>0.128803</td>\n",
       "      <td>0.001897</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3192.025887</td>\n",
       "      <td>0.999899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__254</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__255</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__256</th>\n",
       "      <td>0.949375</td>\n",
       "      <td>0.219231</td>\n",
       "      <td>0.004141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2517.301618</td>\n",
       "      <td>1.000086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__257</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__258</th>\n",
       "      <td>0.000625</td>\n",
       "      <td>0.024992</td>\n",
       "      <td>0.000437</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__259</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__260</th>\n",
       "      <td>0.999687</td>\n",
       "      <td>0.017675</td>\n",
       "      <td>0.000311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__261</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__262</th>\n",
       "      <td>0.003438</td>\n",
       "      <td>0.058529</td>\n",
       "      <td>0.000978</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3224.662207</td>\n",
       "      <td>0.999944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__263</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__264</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__265</th>\n",
       "      <td>0.013750</td>\n",
       "      <td>0.116451</td>\n",
       "      <td>0.002088</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3296.006530</td>\n",
       "      <td>0.999716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__266</th>\n",
       "      <td>0.995938</td>\n",
       "      <td>0.063608</td>\n",
       "      <td>0.001140</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2823.618545</td>\n",
       "      <td>1.000870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__267</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__268</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__269</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__270</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment__271</th>\n",
       "      <td>0.993125</td>\n",
       "      <td>0.082630</td>\n",
       "      <td>0.001295</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3247.614839</td>\n",
       "      <td>0.999916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>centers__0</th>\n",
       "      <td>54.627527</td>\n",
       "      <td>0.733327</td>\n",
       "      <td>0.011941</td>\n",
       "      <td>53.263593</td>\n",
       "      <td>56.069912</td>\n",
       "      <td>2857.383302</td>\n",
       "      <td>1.000063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>centers__1</th>\n",
       "      <td>80.087834</td>\n",
       "      <td>0.523151</td>\n",
       "      <td>0.008877</td>\n",
       "      <td>79.104464</td>\n",
       "      <td>81.134251</td>\n",
       "      <td>3013.404106</td>\n",
       "      <td>0.999852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>p</th>\n",
       "      <td>0.361858</td>\n",
       "      <td>0.030777</td>\n",
       "      <td>0.000567</td>\n",
       "      <td>0.302382</td>\n",
       "      <td>0.419931</td>\n",
       "      <td>2753.669629</td>\n",
       "      <td>1.000464</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sds__0</th>\n",
       "      <td>6.018165</td>\n",
       "      <td>0.585472</td>\n",
       "      <td>0.010211</td>\n",
       "      <td>4.961674</td>\n",
       "      <td>7.231991</td>\n",
       "      <td>2942.626870</td>\n",
       "      <td>0.999759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sds__1</th>\n",
       "      <td>5.947338</td>\n",
       "      <td>0.416616</td>\n",
       "      <td>0.008702</td>\n",
       "      <td>5.151874</td>\n",
       "      <td>6.773165</td>\n",
       "      <td>2566.297885</td>\n",
       "      <td>0.999898</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>277 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      mean        sd  mc_error    hpd_2.5   hpd_97.5        n_eff      Rhat\n",
       "assignment__0     0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__1     0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__2     0.993125  0.082630  0.001504   1.000000   1.000000  3248.753445  0.999745\n",
       "assignment__3     0.050937  0.219870  0.004231   0.000000   1.000000  3173.388070  1.000029\n",
       "assignment__4     1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__5     0.000313  0.017675  0.000311   0.000000   0.000000          NaN  1.000000\n",
       "assignment__6     1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__7     1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__8     0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__9     1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__10    0.000625  0.024992  0.000437   0.000000   0.000000  3208.032629  0.999687\n",
       "assignment__11    0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__12    1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__13    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__14    0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__15    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__16    0.046875  0.211371  0.003711   0.000000   0.000000  3205.553814  0.999696\n",
       "assignment__17    1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__18    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__19    0.999375  0.024992  0.000437   1.000000   1.000000  3208.032629  0.999687\n",
       "assignment__20    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__21    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__22    0.999062  0.030604  0.000533   1.000000   1.000000  3209.379383  0.999792\n",
       "assignment__23    0.817500  0.386256  0.006151   0.000000   1.000000  3120.023664  0.999729\n",
       "assignment__24    0.993125  0.082630  0.001629   1.000000   1.000000  3235.543269  1.001749\n",
       "assignment__25    1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__26    0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__27    0.997188  0.052958  0.001177   1.000000   1.000000  3222.014229  0.999722\n",
       "assignment__28    0.998437  0.039498  0.000681   1.000000   1.000000  3213.694327  0.999750\n",
       "assignment__29    0.998750  0.035333  0.000612   1.000000   1.000000  3212.072926  0.999687\n",
       "...                    ...       ...       ...        ...        ...          ...       ...\n",
       "assignment__247   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__248   0.576562  0.494103  0.008319   0.000000   1.000000  2947.001300  1.000571\n",
       "assignment__249   0.995938  0.063608  0.001140   1.000000   1.000000  2794.058393  0.999712\n",
       "assignment__250   0.000625  0.024992  0.000437   0.000000   0.000000          NaN  1.000313\n",
       "assignment__251   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__252   0.983750  0.126436  0.002048   1.000000   1.000000  3312.296435  0.999907\n",
       "assignment__253   0.983125  0.128803  0.001897   1.000000   1.000000  3192.025887  0.999899\n",
       "assignment__254   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__255   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__256   0.949375  0.219231  0.004141   0.000000   1.000000  2517.301618  1.000086\n",
       "assignment__257   0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__258   0.000625  0.024992  0.000437   0.000000   0.000000          NaN  1.000313\n",
       "assignment__259   0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__260   0.999687  0.017675  0.000311   1.000000   1.000000          NaN  1.000000\n",
       "assignment__261   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__262   0.003438  0.058529  0.000978   0.000000   0.000000  3224.662207  0.999944\n",
       "assignment__263   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__264   0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__265   0.013750  0.116451  0.002088   0.000000   0.000000  3296.006530  0.999716\n",
       "assignment__266   0.995938  0.063608  0.001140   1.000000   1.000000  2823.618545  1.000870\n",
       "assignment__267   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__268   0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__269   1.000000  0.000000  0.000000   1.000000   1.000000          NaN       NaN\n",
       "assignment__270   0.000000  0.000000  0.000000   0.000000   0.000000          NaN       NaN\n",
       "assignment__271   0.993125  0.082630  0.001295   1.000000   1.000000  3247.614839  0.999916\n",
       "centers__0       54.627527  0.733327  0.011941  53.263593  56.069912  2857.383302  1.000063\n",
       "centers__1       80.087834  0.523151  0.008877  79.104464  81.134251  3013.404106  0.999852\n",
       "p                 0.361858  0.030777  0.000567   0.302382   0.419931  2753.669629  1.000464\n",
       "sds__0            6.018165  0.585472  0.010211   4.961674   7.231991  2942.626870  0.999759\n",
       "sds__1            5.947338  0.416616  0.008702   5.151874   6.773165  2566.297885  0.999898\n",
       "\n",
       "[277 rows x 7 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.summary(oftrace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.autocorrplot(oftrace, varnames=['p', 'centers', 'sds']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([54.62752749, 80.0878344 ])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "oftrace['centers'].mean(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can visualize the two clusters, suitably scales by the category-belonging probability by taking the posterior means. Note that this misses any smearing that might go into making the posterior predictive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.stats import norm\n",
    "x = np.linspace(20, 120, 500)\n",
    "# for pretty colors later in the book.\n",
    "colors = [\"#348ABD\", \"#A60628\"] if oftrace['centers'][-1, 0] > oftrace['centers'][-1, 1] \\\n",
    "    else [\"#A60628\", \"#348ABD\"]\n",
    "\n",
    "posterior_center_means = oftrace['centers'].mean(axis=0)\n",
    "posterior_std_means = oftrace['sds'].mean(axis=0)\n",
    "posterior_p_mean = oftrace[\"p\"].mean()\n",
    "\n",
    "plt.hist(ofdata.waiting, bins=20, histtype=\"step\", normed=True, color=\"k\",\n",
    "     lw=2, label=\"histogram of data\")\n",
    "y = posterior_p_mean * norm.pdf(x, loc=posterior_center_means[0],\n",
    "                                scale=posterior_std_means[0])\n",
    "plt.plot(x, y, label=\"Cluster 0 (using posterior-mean parameters)\", lw=3)\n",
    "plt.fill_between(x, y, color=colors[1], alpha=0.3)\n",
    "\n",
    "y = (1 - posterior_p_mean) * norm.pdf(x, loc=posterior_center_means[1],\n",
    "                                      scale=posterior_std_means[1])\n",
    "plt.plot(x, y, label=\"Cluster 1 (using posterior-mean parameters)\", lw=3)\n",
    "plt.fill_between(x, y, color=colors[0], alpha=0.3)\n",
    "\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.title(\"Visualizing Clusters using posterior-mean parameters\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A tetchy 3 Gaussian Model\n",
    "\n",
    "Let us set up our data. Our analysis here follows that of https://colindcarroll.com/2018/07/20/why-im-excited-about-pymc3-v3.5.0/ , and we have chosen 3 gaussians reasonably close to each other to show the problems that arise!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n",
      "//anaconda/envs/py3l/lib/python3.6/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAECCAYAAAD5OrxGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xl8VPW9//HXZCEESCBAIOwhBL4EkN2FRam4gFotLbbu1fvrbfur1Vpttav3d7X74tLWeu1y23uttba1tnUFkeICsmgQFAhfwpIQIAmQhQSyZ+b3x5kMYcwkkzDJTHLez8eDxzBnzvecz5k5+cyZ7/kuHp/Ph4iIuEdctAMQEZGepcQvIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMgnRDiAcubm5anMqItJJ8+bN87S1vFckfoB58+ZFO4SQ8vLyAMjJyYlyJD1Hx9z3ue14oW8dc25ubsjXVNUjIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jK9pueuiJut2lhwxvPlCzKjEYb0EbriFxFxGSV+ERGXUeIXEXEZJX4REZdR4hcRcRklfhERl1HiFxFxGSV+ERGXUeIXEXEZJX4REZdR4hcRcRklfhERl1HiFxFxGSV+ERGXUeIXEXEZJX4REZcJeyIWY8xngfuAscA24B5r7cYwy/4/4D+ttZ4uRSkiIhET1hW/MeZW4AngKWAlUAmsNsZMDKPsDOCbZxOkiIhEToeJ3xjjAR4Afm2tfcBa+zJwDXAcuLuDsvHA74BjEYhVREQiIJwr/mxgAvB8ywJrbSPwErC8g7J3AynAL7oaoIiIRFY4iX+K/3Fv0PL9wCT/Vf2HGGOycX4pfBao73KEIiISUeHc3E31P1YHLa/G+eIYCFS1fsFfPfRb4Elr7XpjzPyzDTQvL+9sN9FtamtrgdiOMdJ0zD2ruKTyjOd5ebXdvk99xn1XOIm/pSWOL8Tr3jaWfR6niuiargQlIiLdJ5zEf8L/mAKUtlqeAjRba0+2XtkYMw74MfBvQI0xJgF/lZL//15rbVtfFu3KycnpbJEe03J1EMsxRpqOuWcVVhac8TwnJ7Pb96nPuHfLzc0N+Vo4dfz5/sesoOVZwJ421r8E50vhWaDR/+8h/2uNwH+EsU8REekm4Vzx5wNFwArgVQBjTCJwFU7LnmAvAOcGLbsBuMe//EhXgxURkbPXYeK31vqMMT8EHjPGVAAbgDuA4cAjAMaYSUC6tXaTtbYMKGu9DWPMYv+23o1w/CIi0klh9dy11j4O3AvcglOFMwRYZq3d71/lfiCs4RtERCS6wh6rx1r7EKfr6oNfuw24rZ2yjwKPdjI26YVWbSwI/H/5gsyw1w1nfRGJDI3OKSLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jJK/CIiLqPELyLiMkr8IiIuo8QvIuIySvwiIi6jxC8i4jIJ0Q5AJByrNha0+/ryBZkR2/bZbEukN9AVv4iIyyjxi4i4jBK/iIjLhF3Hb4z5LHAfMBbYBtxjrd3YzvrLge8A04AjwM+Bx6y1vrOKWEREzkpYV/zGmFuBJ4CngJVAJbDaGDMxxPoLgBeAHcDHgN8ADwNfjkDMIiJyFjq84jfGeIAHgF9bax/wL1sDWOBu4EttFLsb2An8H/8V/mvGmBzgi8AjEYpdRES6IJwr/mxgAvB8ywJrbSPwErA8RJmvADcEVes0AEldjFNERCIknDr+Kf7HvUHL9wOTjDHx1trm1i9Ya4ta/m+MGQJcA3wa+O5ZxCoiIhEQTuJP9T9WBy2vxvnFMBCoaqugMWYCUOB/+i7wX50P0ZGXl9fVot2utrYWiO0YIy3UMReXVAb+n5dX2+42Wq/b0frB6wbraF+RiCOan3Nn3qtI0Xndd4VT1ePxP4ZqjeNtp2wVsBS4EUgDNhpjBoQfnoiIRFo4V/wn/I8pQGmr5SlAs7X2ZKiC1toKYB2AMWYH8D5wLfBkZwPNycnpbJEe03J1EMsxRlqoYy6sLAj8Pycns91ttF63o/WD1w3W0b4iEUc0P+fOvFeRovO6d8vNzQ35WjhX/Pn+x6yg5VnAnrYKGGNWGGPODVq8A2gERoexTxER6SbhJv4iYEXLAmNMInAVsDZEma8DPwladjGQCHzQ+TBFRCRSOqzqsdb6jDE/BB4zxlQAG4A7gOH42+QbYyYB6dbaTf5i3wOeN8b8CvgLTsugB4HXgZcjfRAiIhK+sHruWmsfB+4FbgGeBYYAy6y1+/2r3A9sbLX+Czg9dufi9OC9H/gDcJWGbBARia6wx+qx1j4EPBTitduA24KWPU+rTl8iIhIbNDqniIjLKPGLiLiMEr+IiMso8YuIuIwSv4iIyyjxi4i4jBK/iIjLKPGLiLiMEr+IiMso8YuIuIwSv4iIyyjxi4i4TNiDtIm0ZdXGgrBfW74gsxsjCV97MffkfmPl/QjWEmfLPL99YDIqCaIrfhERl1HiFxFxGSV+ERGXUeIXEXEZJX4REZdR4hcRcRklfhERl1HiFxFxGSV+ERGXUeIXEXEZJX4REZdR4hcRcRklfhERl1HiFxFxGSV+ERGXUeIXEXEZJX4REZdR4hcRcRklfhERl1HiFxFxmbAnWzfGfBa4DxgLbAPusdZubGf9hcD3gDlADfAacK+1tvSsIhYRkbMS1hW/MeZW4AngKWAlUAmsNsZMDLF+DrAWqAZuAL4KLPKXSYxA3CIi0kUdXvEbYzzAA8CvrbUP+JetASxwN/ClNordARQDK621jf4y+cAW4DLg5YhELyIinRbOFX82MAF4vmWBP5m/BCwPUWYn8FBL0m8p5n9s81eCiIj0jHDq+Kf4H/cGLd8PTDLGxFtrm1u/YK19vI3tXO1/3N25EEVEJJLCSfyp/sfqoOXVOL8YBgJV7W3AGDMO+CnwLvCvTsYIQF5eXleK9Yja2logtmOMtJZjLi6pDLtMXl7tGc+Dywa/3t66HW27PV3d1tl+zp053kiW7ayWfTU2NPr35b7zuq8fczhVPR7/oy/E6972CvuT/lr/vq631obajoiI9IBwrvhP+B9TgNZNMVOAZmvtyVAFjTEzgFeAROAya+2+rgaak5PT1aLdruXqIJZjjLSWYx6VMTTsMjk5mWc8L6wsaPf19tbtaNvt6eq2zvZz7szxRrJsZ7Xsq7ik2L8v953XfeGYc3NzQ74WzhV/vv8xK2h5FrAnVCFjzPnAW0AzcKG19v0w9iUiIt0snCv+fKAIWAG8CuBvi38VTsueD/G3738FKAEusdYeiUi0Il2wamPBWa2/fEFm4P+bdleecfXd+jWR3qLDxG+t9Rljfgg8ZoypADbgtNMfDjwCYIyZBKRbazf5iz2Kc1P4i8B4Y8z4VpsstNYWR/AYRESkE8LquetvnnkvcAvwLDAEWGat3e9f5X5gIwR+DVwJxANP+5e3/ndTBOMXEZFOCnusHmvtQ8BDIV67DbjN//9GnJu5IiISg8JO/CI9panZS1FpNfsOneBgaTVlJ2rZd6iShkYvzV4fPp+PhPg4EhLiSO6XwID+CVTXNDB2RArjM1LIGDqA+HgNPCsSihK/xISGxmZe21LIll2lbNtzjNr6pk6V37G/LPD/hPg4xqQPZOKYwUybOIyKqjrSUpLweDztbEHEPZT4JarKTtTy/t7j7DlYSVNzu30BAYjzePB4oNkbuh9gU7OXwpJqCkuqeT33EAD9+8WTMWwg40YOInPUYFIH9ovYMYj0Nkr8EhVHjp/k1c2F5Bd9ePiEgf0TmDk5neyxQ5g4OpURQwfwnj1KUmJ84Krd5/PR2OSltr6JmromMkenUlRazcGSag4drab4+ClafzfUNTRTUFxFQXEVb207wtDU/mSOSiVzVCoZwwbo14C4ihK/9KjGJi9/XbuHv7y254yr9oT4OC6eN5aL548jJ3MoCUF19HkHys947vF46JcYT7/EeAYPSuLieePOeL2+sZn8gxXsPFDGm1sPU1x2isam078oyqvqKK+qY6s9yqDkRCaPG8Lk8WkMH9xfXwLS5ynxS4/Zd6iSR/60lcKS0+P9JSXGM8ekM2PScD520aSI7SspMZ4Zk4YzY9JwBg9MwuvzcayiloLiKgqLqzhWeXqQs5O1jby35xjv7TlGWmoSU8alMXVCGoMGqDpI+iYlfukRuwvK+fU/Pghcdcd54JzsdObnjKB/v+4/DeM8HkYOHcDIoQM4f3oGJ2sbKSiuYv+hSg4dPRkYgbCiqp7NO0vYsrOECaNSmTZxKM3NXrUSkj5FiV+6ldfrY/32w3yw73Srm/EZKdx13Rz2Hz7RTsnuNSg5kRlZw5iRNYxTdY3sLapkz8FKjlbUAM5QtC33BDbtKOHy8ycwaXjnWhqJxColfuk2Tc1eXt1cyIEjp6druPTc8Xxh5Uz6JcZHNfG3NrB/IrMmpzNrcjqV1fXkFZSzu7Ccmjon0ZdX1fHMGkucB0YPT+L8xFQyhg2MctQiXafEL92iobGZl98+wOFjpwCnaufzn5jJFQsyY/rm6ZCUJBacM4rzpmdQWFzFzgNlFJVW4/OB1weHjtVzaN1eRg4dwMzs4Vxy7ngSE1QNJL2LEr9EXGNTMy+u309xmVNtkhDvYdkFmVy5sPdMtxwf5yFrzGCyxgxmjhnByxsO8Mrb+6ltcO5RlJbXsGbLQXJ3l3LlwolcsXCi+gZIr6HELxHV1OzllbcLAkm/X2IcVy2ayOjhg6IcWdeNHDqAf7t6OnMnwLPrSyk82kBFdT0A5VX1PLVqN3/9Vz6Xnz+BFRdNYsTQAVGOWKR9SvwSMV6fj1c3F1J01JmULTEhjqsXZ/WZ+vB+iXFkjUpm4eyJHDp6ku35xwJNU+sbmnnhrf28tOEAF80ewycuzmbi6MFRjlikbUr8EjFvbz8SuJEbH+fhqkUT+0zSb83j8TBuZArjRqYwM3s4f39jH2vfOUhjkxev18frWw/x+tZDzJs6gpVLJzMja1hM39cQ91Hil4j4YN9xtu89DoAHZ2aqMem9t3onXKPTB/HFa2dx4+WGF9bv5+UNBzjlbw2Uu/soubuPYsan8YmLs/H5fPoCkJigxC9nrai0mre2HQ48v3D2GDJHpUYxop6XltqfT185jWuXTmb1pkL++eY+yk7UAWAPVvCD/32HtJQk5ueMJHvcEOL0BSBRpHZoclaqTjXw6uZCfP6urzOzh3NO9vDoBhVFA/on8vGPZPObb17GXdfNZuyI0796KqrrWbPlIH961WIPVrQ7wqhId9IVv3RZY5OXVZuKqGtoBmDciEEsmjU6ylHFhsSEOC49bwJL549ny64Snl2bjz1YAUBldT2vbTlI3oFyrr9sChfOGUt8nH4BSM/RFb902fObjnGswhnsbNCARC47f4KqMILExXm4YMYofvKlC7l68URGtmrqefjYSR56eitf/PFa1uUW6ReA9Bhd8UuXfHCgms27nSEX4uM8XHFBJslJOp1C8Xg8jM9IZdzIFIpKq9myq5TScqevw+Fjp3j46a08t24vn74yh/k5I3UTWLqV/lKl08pO1PK39aWB54tmjVanpTC1/gIYOWwgT6/ejS10qoAKiqt48L83Mz1rGLddNY2pmUOjHK30VUr80iler4+Hn95KTb0zdEHmqFRmZA2LclS9j8fjYa4ZwZwp6Wy1R3nypTz2H3F+Qe3cX8a9v3iL86dn8Okrcxif4a4WUtL9lPilU/7xxl7e97fXT0mOZ+n8caqWOAsej4d5U0cyZ8oI3tx2mKdeyQtUAW3eWcI7u0pYOn88Y0YMYlByYpSjlb5CiV/CtvdQJX94JS/w/FNLMuivev2IiIvz8JG5Y1k0czSrNxXw5zV7qDxZj9cHr71zkIR4D7OnjGCOSadfQny0w5VeTq16JCx19U389KlcmpqdlieLZwzBjO17wzFEW2JCHB9dnMWvvnEJNy6bSnKSk+Sbmn28m1fKH1ftZteBMrUAkrOiy7U+aNXGgjOeL1+Q2eXyLWV//+JODh9zBl/LHJXKFfM730krOK7u1JP76up+O1r3hssNVyzI5OnVu1m1sQAfUFPXxLrcQxQUV/GZq2cwa0p6WNtu7xzoKI62zodwt9XZc096hq74pUMf7D3Oy28XANAvIY6v3jxPk4/0kCEpSdx+7Syuu8wwPiMlsPzAkSq+/au3efC/N1FUWt3OFkQ+TH+90q66+iZ+/pf3As9vWp7DBLUy6XHDBvfn6sVZXL04i6Gp/QPL39lVyh0/XccTz73PiZP1UYxQehNV9Ui7nnwljxL/pCpmfBofWzIpyhG52/iMFMaOnEJ8nIenVu2msroer9fHSxsOsC63iOsuncJHF2dFO0yJcUr8EtKR4yd5cf1+ABLi4/jSdbM1pkwMiPM4U1leOHsMz63by99f30tDk5eauiZ+/+IuXnq7gNmTh5M9doia2kqbVNUjbWps8vKvd4sCo27euMyoI1GMGdA/kZuvyOGJr1/KxfPGBpYfLa/h1c0HeW7dXkrKTkUxQolVSvzSpi07SzhxsgGA7HFD+MRHsqMckYSSnpbMPTfO4+EvX8T0Vr2oS8pr+Nu6vazeVBjoFCYCSvzShpKyU2zLPwZAQryHL183h/h4nSqxbvK4NH5w+yK+edu5DB7UL7B876FKvvCjtfzPizs5VdsYxQglVoRdx2+M+SxwHzAW2AbcY63dGEa5FGAH8BVr7bNdDVR6RkNjM/96tyjw/LrLDBNcNptWb+bxeFhwzmjKq+rYsa+Md3aVUt/YTGOTl7+t28uaLQe5cdlUll8wQV/mLhbWJ2+MuRV4AngKWAlUAquNMRM7KJcC/BMYf5ZxSg95evVuKqqdZoHDh/Tn2qWToxyRdEV8XByzJqdz8xVTmZU9PHBTvupUA0889z53PrSOd/NKO9iK9FUdJn5jjAd4APi1tfYBa+3LwDXAceDudsotAbYAsyMUq3SzPQcr+PvrewGI88DS+eNJ0FVhr9a/XwKLZ4/h8fuWsuCcUYHlRaUneeC3m3jgt5uorFb7f7cJ5686G5gAPN+ywFrbCLwELG+n3D+ADzpYR2JEY1MzP/vze7QMATN36kjShyRHNyiJmNHpg/jmbefxg9sXkT12cGD5u3ml/OlVy8YPjtDQ2BzFCKUnhZP4p/gf9wYt3w9MMsaEGirwQmvtp4CjXQ1Oes4za/ZwsMTp+j9scH/m54yIckTSHWZMGs5Ddy3hruvmMCQlCQCvz8dWe4w/+ieF8fk0AFxfF87N3ZY7e8EDglTjfHEMBKqCC1lrd5xdaGfKy8vreKUoqa115p2NlRiLSyrPeJ6XV9vu+oeO1/Hs2oOAU8Uzc2IyR4+Wtlu25ZiD93U22ouzo/0El+2OuGpra2lsaKS4pLhH99vWdjv6TDtaf0wK3PPxsbz2Xjlv7ajA53MGgHvtnYNs3V3MrEmDaGkY1N7xdna/sS7W/pa7SzhX/C1d/0JdBngjFItEQVOzj7++WRKo4lkycyhpgzThhxv07xfPR89P59I5Qxkx5HTzz/LqRtZtq2DnwZrAMNzSt4RzxX/C/5gCtG4GkAI0W2tPRjyqNuTk5PTEbrqk5eogVmIsrCw443lOTmbIdZ9alUdxudNRa9zIFO68cSFr3zndnDNU2ZZjHpURuXlh24sz+Jg6KtvR+p3Rsu28vDwS+9UzKmPUh17rzv22td323qvOrl9YWcDkiT4KiqtYv/0IVaec82F/ST2llY1cel4mY0ekRHy/sSjW/pbPRm5ubsjXwrniz/c/Bo/8lAXs6WJMEgP2HKzgr2udjzfOA1++fg6Jmt3JlTweDxNHD+aGyw3nThtJnH+Mn1N1Xv755n7W5RZR36Cbv31FuIm/CFjRssAYkwhcBaztprikmzU0NvPoM1vx+ut4Vi6dzJTxaVGOSqItIT6O86Zl8KlLp5CWcrpCYNeBcv60xvKBf75l6d06TPzWWh/wQ+D/GmO+Z4y5EqdT1nDgEQBjzCRjzAXdGqlE1FOrdlNUenpGrRsuN1GOSGLJsMH9+cjMNGZOHBToy3GqtpFvPbGBJ1/eRVOzbu31ZmH1zrHWPg7cC9wCPAsMAZZZa/f7V7kf6HD4BokNuw6U8Y83nNa58XEe7r5hrqp45EM8Hg/ZYwZw/WVTGDVsAAA+H/x1bT5f/+V6jfzZi4U9Vo+19iHgoRCv3QbcFuK1Ak63DJIoq6tv4tFn3gsMt3z95YasMYPbLySuNnhQEiuWZPPu7lJy80rx+sAWVnDXw69zxydnc+HsMdEOUTpJ/fFd5n9f2kXxcedKLXvsYI3FI2GJi/Nw3rQMvn/7YtLTnB7dNXVN/PgP7/K7F3bSrKqfXkWJ30XezSvlxQ0HAEhMiOPLN8zVWDzSKdOzhvHzez7CopmjA8v+/vpe/vM3mwLNQCX26a/eJcqr6nj0ma2B57dcoUnTpWsGDejH1z49n89cM52WmTi35R/j7kff4Fhl7+qp61ZK/C7g9fp45OmtgRm15kxJ52MXadJ06TqPx8OKJdk8+LmFpAxwev0eLa/huXX55BdFbtgK6R5K/C7wjzf2BmbUGjIoibtvmEucJk2XCJg1JZ1H7l5C1mingUBTs49XNxeSu7tUg73FMCX+Pq60vIYnXz494NRd188hLbV/FCOSvmbk0AH86M7FXDTndOueTTtKWJd7SO39Y5QSfx/W0NjMq5sLafb3zv3YRZOYnzMyylFJX9S/XwJfvWneGedXXkE5//mbjZzUPL8xR4m/j/L5fLy+9VCgpUXWmMHcelXvH3hKYpfH4+H86Rlccu64wFg/2/OPc98v3uJoeU2Uo5PWlPj7qPf3Hg/cZEvqF8+9N89T71zpEVMnDOWaC7NISnTOt6LSau79xZscOHKig5LSU8LuuSuxbdXGgsD/Dx09yYb3jwSeXzR7DDv2lbFjX1nPB9YHtLy3bU2y0vp9j6bOxBHJmENta8yIQay8OJsXNxyg6lQD5VX1fOVnb3LFgkzGjUxps8zyBZlh76ujdaV9uuLvY6pONbB6U0FgSIZZk9M16qZERVpqf1ZenB3o6dvY5OXF9fuxhRVRjkyU+PuQpmYvqzYWUOcfN31M+iAWnjOq/UIi3WhA/0RWLJnE+AznKt/rg9feOajmnlGmxN9H+Hw+Xs89FOg5mTIgkWUXTFB7fYm6fgnxXLlwIlMzT//y3LSjhDffO4xXyT8qlPj7iO35x7EHnZ/Q8XEerliQSXKSbuFIbIiP87B03jjObdXcc8f+MlZtLFBb/yhQ4u8DNrx/5IybuRfPH0d62oAoRiTyYR6Ph/OmZ3DxvLH4W3ty4EgV/3xjH7X1TdENzmWU+Hu5vAPlPPzH05MqzzUjMLqZKzFs2sRhXLlwYmBk2JLyGp5bt1cTu/QgJf5erLCkiu/8bhMNTc5P5cnjhnDBjIwoRyXSscxRqaxYMilQHVl5sp57f/4WezXAW49Q4u+lSspO8R+/epvqGqc7/KjhA7lk/jg8Ht3Mld5h5NABrLw4m8GDnNE9K0/W843H1/NuXmmUI+v7lPh7obITtdz/q7cpr6oHnKunKxdmEq9JVaSXGTwoiZUXT2aE/55UXUMz3/ndZl7acEDNPbuRMkUvc7yylm88voGSMmfsk1HDB/Lg5xfQv59a8EjvlJyUwIolWZw7zWnx4/X6eOK59/nls9tpbFKLn+6gxN+LHKuo5ZuPbwjMmZuelsx3P7+QtBQNsyy9W2JCPN+67Tw+umhiYNnqTYV8+4kNVFbXRzGyvkmJv5c4dLSar/3yLYr9LR9GpCXz/S8sYsRQNduUviE+Po7Pf2Imd3xyNgnxzr2qXQfKufvRN9h3SDd9I0mJvxfIL6rga4+t51iF0yt3xNABfP/2xWQMGxjlyEQib9kFE/jeFxYxJCUJcKo37/vFW+zYd1z1/hGixB/jNu8o5puPbwiMqz9uZAo/+uJiRupKX/qwaROH8fBdS8ge60zp2NDk5Y33DrNqYyF1DersdbaU+GOUz+fjuXX5fO9/tgQGXZs6IY0f3bGY4UOSoxydSPdLT0vmh3dcyOXnTwgs23/kBH9es4cP9h2PYmS9nxJ/DKqpa+Snf8zl9y/uCgyvvOCcUXzn8wtJGdAvusGJ9KCkxHju/NRsvvbp+fRLdNLVydpGvv1fG3jy5V00NDZHOcLeSW0AY0xhcRU/+N93OHzsZGDZJy+ZzM3LczTSprjW4lljOHLsFGu2FFJSVoPXB39dm8/67Uf4widmMseMiHaIvYoSf4xo9vp4/s19/OGVvEDb5f794rnjk7NZMndslKMTib7Ugf34+JJs3skrZas9itfro/j4Kf7j1xu5aM4Y/v2aGaSlqmlzOJT4Y0BRaTW/fHY7O/efnhpx3MgUvnHruSGnqRNxo7g4Z0L3W67I4ZfPbmPPQaeZ55vvHSY3r5Qbl03lioWZml+6A0r8UVRX38Rf1u7h76/vpan5dDO1qxZN5LarptFf4+mLtClrzGB+fOdFrN5UwJMv7eJUXROn6pr4zT938Pxb+7lx2VSWzBmjYUxCUGaJgqZmL69uLuSZVy0VrXolDh+SzF3XzWb2FNVXinQkPs7DlQsnsmDGKH77zx28ue0wAKXlNTzyp60886pl5dLJLJ0/Vr8Agijx96CGxmbWvnOQ517fGxhrB5wTeMWSSVx3mdGsWSKdlJban3tvmc/VF2XxPy/uClSZFped4rG/buMPr+xi+QWZLLsgMzDxu9spy/SA0vIaVm8qYM3mg1SePHPckfOmZXDrVTmMz0iNUnQifcPUCUP5we2LeG/PMf68xrLrQDkAJ0428OfX9vCXtXuYnjWMJXPGsmjWaFc3jQ478RtjPgvcB4wFtgH3WGs3trP+DOBnwPlAOfBL4MfWWlf0ua461cDGD47w5nuH+WDfcYJ7ms+YNIybl+cwPWtYdAIU6YM8Hg9zzQjmmhHs2Hec59/az+YdxXh94PPM6XokAAAKaUlEQVTBjn1l7NhXxq/+/j7zpo7k3GkjmTU53XXDn4SV+I0xtwJPAA8C7wB3AquNMbOstQfaWH8E8BqwA/gUMBf4HtAM/DQyoceWhkYvO/YdZ+eBMnLzjmILy/EGJfs4D5w3PYOVSyczdcLQ6AQq4hIzJg1nxqThlJbXsGZzIW+8dyhQxdrU7GPzzhI27ywBnElhZk9JZ2hyPWOH98d4fX2630yHid8Y4wEeAH5trX3Av2wNYIG7gS+1UeyL/m1fY62tAV42xiQB3zDG/Mxa2xipA+hpPp+Pyup6jhw/xZFjJ8kvqmT7nmJKyuvx+va2WSY9LZlL5o/n8vMnqI5RpIeNHDqAm6/I4ablU7EHK3hj6yHWbztyRrWrUx1bGHj+2PNFZI4azKQxgxk/KpVRwwYwcuhA0tOSA3MF92bhXPFnAxOA51sWWGsbjTEvActDlLkUWOtP+i3+AXwbOBd4u2vhRobP58Pr9dHk9dHc7KWxyUtdQzO19U3U1jVRW9/EqbpGKqvrOXGynsqT9VRW13OsopbispPU1nfcTTxj2ADOnZbBhbPGYCak9emrB5HewOPxMHXCUKZOGMq/XzODvIJytucfZ3v+MezBCrytfqLX1jeTV1BOXkH5GduI8zit74YNTmZIShKDByUxeFA/Bg9MYmByAslJCfTv5zwmJyXQPymB/v3i6ZcYT0K8h7i4OOJjIBeEk/in+B+DL2f3A5OMMfHW2uBMOAV4vY31W17r9sR/vLKWnz3zHgdLq2hqdhJ8S6Jv3WY+UkYM7seMySMxE9KYNTmd0cMHav5bkRgVHx8XqAq6aflUauoa2bG/jA3v5nO4rI5jVc2BqU1b8/rgaEUtR/1DpHeFx+O05IuPjyMhzvkySIj3OF8I/pzh8YAHSE8bwBevnRXxjpzhJP6W5ibVQcurcQZ5GwhUtVGmrfVbb69T8vLyOrX++h0VbMs/1pVdheQBhgxKYFhqP4alJjI8NZGMtCTSUyA5KY7k5GSgjqrjRVT18OCBxSXdM1FFXl7bJ3htbW3E9xtqX+HsJ7hsd7wfjQ2N/m0XR3zbbWl9TMHH09njbW9bobR1vN35Prf3+Qfvq6N1uyLFA0tmDAQGkpycTHVNE0fK6zl+ooGy6kYqqhspr26krLqRhsauXzz6fM49hqbmZjqaW+xoRS1/WfUeV56X3uX9tSWcxN9y2RrqSNuaFNPTyfU7VFNT0/FKrczNSmJuVs+NcePz+TodYyTNHN89TdM6OqZI7re9fXW0n+Cy3fN+9Gzzv9bHFHw8nT3e9rYV2ofX6873uTPnWnf/rdXU1BAPjBvqYdzQJCCpW/cXTjyRFE7iP+F/TAFKWy1PAZqttSc/XIQT/tcJWr/19sI2b9481ZmIiERIOLen8/2PWUHLs4A97ZRpa31wWgOJiEiUhJv4i4AVLQuMMYnAVcDaEGXWApcaY1r3ilgBlOF0/hIRkSjxhDN5sTHmduAx4AfABuAOYDEw21q73xgzCUi31m7yrz8KyAO2Az8BZuH0Bfi6tbZPduASEektwuqJYK19HLgXuAV4FhgCLLPWtjTRvB/Y2Gr9Ypy2/An+9T8HfEtJX0Qk+sK64hcRkb6j9/c9FhGRTlHiFxFxGSV+ERGXUeIXEXEZzcDVBcaYF4CPtvFSSoiezPiHpf4hcAPO+EargS9Za490W6ARZIxZiDOnwhygBme+hXuttaXtlFmJ06or2J3W2se6JdCz4LbJhowx8cBdwGeB8UAh8Djwy1DH0JVzP5YYY4YBbY2k9Tdr7bUhyvTqz7ktSvxdMwvnRHgmaHl7A2o8AVwDfAU4idMn4mVjzLw2RjeNKcaYHJxOeWtwvrjSgO/gTMZzbjvzK8zCGdX1lqDlH5q8J9pcOtnQ/cDXcT7LTcCFwKPAAODHIcp05dyPJbP8j5dz5kCSZW2t3Ec+5w9R4u8kY8wQYBywqqXDWhhlJgGfBm601v7Zv2w7zvAVHwOe66ZwI+UOoBhY2ZLkjTH5wBbgMuDlEOVmArnhvk/R4sbJhvxX+/cAP7HWfs+/eK0xJh34Km0k/q6c+zFoJlBqrV0T5vq9+nMORXX8nTfT//h+J8os9T++2LLAWpsP7CT0ZDaxZCfwUNBJ3jLm0sR2ys2kc+9TtLQ52RDQlcmGhuJMNhTrUoEn+fBFhwXSg4ZbadGVcz/WdPac7O2fc5t0xd95M4F64LvGmI8ByTgJ4k5rbUmIMlOAEmvtqaDl+zk90U3M8vfcDna1/3F3W2WMMSlAJjDHGLMH5wsiD2fYjlC/EKKlV042dDastRU4v+SCXQ0cauNcha6d+7FmJlBnjHkbp9rmOE7V1U9D1Nn36s85FCX+VvyDz01qZ5VSnBMnCad+8OM4o45+F/iXMWaOtbatuRXampgG/7JxZxX0WQrnmP1JonWZcTj1m+8C/wpR7hyceRkm4lQpNAG3Ay8YYy611q4729gjKCYmG4o2Y8y/41zhtlW1BV0792OGv3prGnAKpzqrEGewyR/ifIk92EaxPvc5gxJ/sDE4V6Wh3A08DPypVeJ60xiTh3Nz7FPAH9ooF/GJaSIonGN+tOWJP+mvxUmI17fTsmEXzh/Vemttlb/sGpyB+74NxFLij4nJhqLJGHMTzs3tZ3EGZGxLV879WPNR4KC1tuXX3evGmEHA14wxP7bW1gWt36c+5xZK/K1Yaws4nQTac0b1hrV2szGmEqfFQFsnf1sT0+Bf1umJaSKpE8fc0qztFSARuMxau6+d7VYSdNPXWtvsT/7BrXyiLeqTDUWTMeYenF9wzwM3hfoyt9bupvPnfszwV9e19Qt1FfB/ce717Ah6rc98zq3p5m4nGWOuN8ZcFLTMg/MTONRMu/lAhjEmOWh5Fr1kYhpjzPnAWzjN2C601rZ7g8wYM8dfdRAsmdDvU7S4drIhY8z3gYdwkva11tqGdtbtyrkfM4wxo40xn/O3XGqt5e+yrWPoE59zMCX+zvsC8DNjTOv37kqck+fNEGXWAvGcviGKMWYyMJ3Qk9nEDGPMRJwr/RJgob9FUkdmA78xxsxptZ1knPfqjW4JtOtcOdmQMeYu4Bs4Nzdvs9Y2dVCkK+d+LEkCfgXcHLR8JbAnxA3qXv85t0XDMneSMWYZThJ8Bvg9zp397wD/aun5Z4xJxbmJtM9ae8y/7C/AMpybShU4HbhOAb2hA9c/cZLgLXy481WhtbY4+Jj99aZbcS4uvgXU4szpMB2YZa0t6rEDCIPbJhvyx38A5xfN59pY5V2cJq6tj7nDcz/WGWOexulI+S2cz++TwGeAFdba5/va5xyKrvg7yVq7GufEycZpz/st4HecWW89F2dimqtaLfs34M/Aj4Df4pxIV/aCpJ+Ic1UXDzyNc1yt/93kX/WMY/bXi1+C0wv258CfcHp3XhRrSR9cOdnQMpwr4HP48Ge6Eef4g485nHM/1n0G53z8Ms49jfk4HRNb+nD0tc+5TbriFxFxGV3xi4i4jBK/iIjLKPGLiLiMEr+IiMso8YuIuIwSv4iIyyjxi4i4jBK/iIjLKPGLiLjM/wfGVXAcOgrCaQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mu_true = np.array([-2, 0, 2])\n",
    "sigma_true = np.array([1, 1, 1])\n",
    "lambda_true = np.array([1/3, 1/3, 1/3])\n",
    "n = 100\n",
    "from scipy.stats import multinomial\n",
    "# Simulate from each distribution according to mixing proportion psi\n",
    "z = multinomial.rvs(1, lambda_true, size=n)\n",
    "data=np.array([np.random.normal(mu_true[i.astype('bool')][0], sigma_true[i.astype('bool')][0]) for i in z])\n",
    "sns.distplot(data, bins=50);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.savetxt(\"data/3gv2.dat\", data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "with pm.Model() as mof:\n",
    "    #p = pm.Dirichlet('p', a=np.array([1., 1., 1.]), shape=3)\n",
    "    p=[1/3, 1/3, 1/3]\n",
    "\n",
    "    # cluster centers\n",
    "    means = pm.Normal('means', mu=0, sd=10, shape=3)\n",
    "\n",
    "\n",
    "    #sds = pm.HalfCauchy('sds', 5, shape=3)\n",
    "    sds = np.array([1., 1., 1.])\n",
    "    \n",
    "    # latent cluster of each observation\n",
    "    category = pm.Categorical('category',\n",
    "                              p=p,\n",
    "                              shape=data.shape[0])\n",
    "\n",
    "    # likelihood for each observed value\n",
    "    points = pm.Normal('obs',\n",
    "                       mu=means[category],\n",
    "                       sd=1., #sds[category],\n",
    "                       observed=data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n",
      "CompoundStep\n",
      ">NUTS: [means]\n",
      ">CategoricalGibbsMetropolis: [category]\n",
      "Sampling 2 chains: 100%|██████████| 21000/21000 [01:39<00:00, 210.02draws/s]\n",
      "The estimated number of effective samples is smaller than 200 for some parameters.\n"
     ]
    }
   ],
   "source": [
    "with mof:\n",
    "    tripletrace_full = pm.sample(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_base.py:3604: MatplotlibDeprecationWarning: \n",
      "The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
      "  alternative='`bottom`', obj_type='argument')\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x144 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trace_mof=tripletrace_full[3000::7]\n",
    "#pm.traceplot(trace_mof, varnames=[\"means\", \"p\", \"sds\"]);\n",
    "pm.traceplot(trace_mof, varnames=[\"means\"], combined=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.autocorrplot(trace_mof, varnames=['means']);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problems with clusters and sampling\n",
    "\n",
    "Some of the traces seem ok, but the autocorrelation is quite bad. And there is label-switching .This is because there are major problems with using MCMC for clustering.\n",
    "\n",
    "AND THIS IS WITHOUT MODELING $p$ OR $\\sigma$. It gets much worse otherwise! (it would be better if the gaussians were quite widely separated out).\n",
    "\n",
    "These are firstly, the lack of parameter identifiability (the so called label-switching problem) and secondly, the multimodality of the posteriors.\n",
    "\n",
    "We have seen non-identifiability before. Switching labels on the means and z's, for example, does not change the likelihoods. The problem with this is that cluster parameters cannot be compared across chains: what might be a cluster parameter in one chain could well belong to the other cluster in the second chain. Even within a single chain, indices might swap leading to a telltale back and forth in the traces for long chains or data not cleanly separated.\n",
    "\n",
    "Also, the (joint) posteriors can be highly multimodal. One form of multimodality is the non-identifiability, though even without identifiability issues the posteriors are highly multimodal.\n",
    "\n",
    "To quote the Stan manual:\n",
    ">Bayesian inference fails in cases of high multimodality because there is no way to visit all of the modes in the posterior in appropriate proportions and thus no way to evaluate integrals involved in posterior predictive inference.\n",
    "In light of these two problems, the advice often given in fitting clustering models is to try many different initializations and select the sample with the highest overall probability. It is also popular to use optimization-based point estimators such as expectation maximization or variational Bayes, which can be much more efficient than sampling-based approaches."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Some mitigation via ordering in pymc3\n",
    "\n",
    "But this is not a panacea. Sampling is still very hard.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "import theano.tensor as tt\n",
    "import pymc3.distributions.transforms as tr\n",
    "\n",
    "\n",
    "with pm.Model() as mof2:\n",
    "    \n",
    "    p = [1/3, 1/3, 1/3]\n",
    "\n",
    "    # cluster centers\n",
    "    means = pm.Normal('means', mu=0, sd=10, shape=3,\n",
    "                  transform=tr.ordered,\n",
    "                  testval=np.array([-1, 0, 1]))\n",
    "\n",
    "\n",
    "                                         \n",
    "    # measurement error\n",
    "    #sds = pm.Uniform('sds', lower=0, upper=20, shape=3)\n",
    "\n",
    "    # latent cluster of each observation\n",
    "    category = pm.Categorical('category',\n",
    "                              p=p,\n",
    "                              shape=data.shape[0])\n",
    "\n",
    "    # likelihood for each observed value\n",
    "    points = pm.Normal('obs',\n",
    "                       mu=means[category],\n",
    "                       sd=1., #sds[category],\n",
    "                       observed=data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n",
      "CompoundStep\n",
      ">NUTS: [means]\n",
      ">CategoricalGibbsMetropolis: [category]\n",
      "Sampling 2 chains: 100%|██████████| 21000/21000 [03:42<00:00, 94.44draws/s] \n",
      "The acceptance probability does not match the target. It is 0.9255675538672136, but should be close to 0.8. Try to increase the number of tuning steps.\n",
      "The number of effective samples is smaller than 10% for some parameters.\n"
     ]
    }
   ],
   "source": [
    "with mof2:\n",
    "    tripletrace_full2 = pm.sample(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_base.py:3604: MatplotlibDeprecationWarning: \n",
      "The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
      "  alternative='`bottom`', obj_type='argument')\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x144 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trace_mof2 = tripletrace_full2[3000::5]\n",
    "pm.traceplot(trace_mof2, varnames=[\"means\"], combined=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.autocorrplot(trace_mof2, varnames=[\"means\"]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Full sampling is horrible, even with potentials\n",
    "\n",
    "Now lets put Dirichlet based (and this is a strongly centering Dirichlet) prior on the probabilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from scipy.stats import dirichlet\n",
    "ds = dirichlet(alpha=[10,10,10]).rvs(1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Visualize points on the 3-simplex (eg, the parameters of a\n",
    "3-dimensional multinomial distributions) as a scatter plot \n",
    "contained within a 2D triangle.\n",
    "David Andrzejewski (david.andrzej@gmail.com)\n",
    "\"\"\"\n",
    "import numpy as NP\n",
    "import matplotlib.pyplot as P\n",
    "import matplotlib.ticker as MT\n",
    "import matplotlib.lines as L\n",
    "import matplotlib.cm as CM\n",
    "import matplotlib.colors as C\n",
    "import matplotlib.patches as PA\n",
    "\n",
    "def plotSimplex(points, fig=None, \n",
    "                vertexlabels=['1','2','3'],\n",
    "                **kwargs):\n",
    "    \"\"\"\n",
    "    Plot Nx3 points array on the 3-simplex \n",
    "    (with optionally labeled vertices) \n",
    "    \n",
    "    kwargs will be passed along directly to matplotlib.pyplot.scatter    \n",
    "    Returns Figure, caller must .show()\n",
    "    \"\"\"\n",
    "    if(fig == None):        \n",
    "        fig = P.figure()\n",
    "    # Draw the triangle\n",
    "    l1 = L.Line2D([0, 0.5, 1.0, 0], # xcoords\n",
    "                  [0, NP.sqrt(3) / 2, 0, 0], # ycoords\n",
    "                  color='k')\n",
    "    fig.gca().add_line(l1)\n",
    "    fig.gca().xaxis.set_major_locator(MT.NullLocator())\n",
    "    fig.gca().yaxis.set_major_locator(MT.NullLocator())\n",
    "    # Draw vertex labels\n",
    "    fig.gca().text(-0.05, -0.05, vertexlabels[0])\n",
    "    fig.gca().text(1.05, -0.05, vertexlabels[1])\n",
    "    fig.gca().text(0.5, NP.sqrt(3) / 2 + 0.05, vertexlabels[2])\n",
    "    # Project and draw the actual points\n",
    "    projected = projectSimplex(points)\n",
    "    P.scatter(projected[:,0], projected[:,1], **kwargs)              \n",
    "    # Leave some buffer around the triangle for vertex labels\n",
    "    fig.gca().set_xlim(-0.2, 1.2)\n",
    "    fig.gca().set_ylim(-0.2, 1.2)\n",
    "\n",
    "    return fig    \n",
    "\n",
    "def projectSimplex(points):\n",
    "    \"\"\" \n",
    "    Project probabilities on the 3-simplex to a 2D triangle\n",
    "    \n",
    "    N points are given as N x 3 array\n",
    "    \"\"\"\n",
    "    # Convert points one at a time\n",
    "    tripts = NP.zeros((points.shape[0],2))\n",
    "    for idx in range(points.shape[0]):\n",
    "        # Init to triangle centroid\n",
    "        x = 1.0 / 2\n",
    "        y = 1.0 / (2 * NP.sqrt(3))\n",
    "        # Vector 1 - bisect out of lower left vertex \n",
    "        p1 = points[idx, 0]\n",
    "        x = x - (1.0 / NP.sqrt(3)) * p1 * NP.cos(NP.pi / 6)\n",
    "        y = y - (1.0 / NP.sqrt(3)) * p1 * NP.sin(NP.pi / 6)\n",
    "        # Vector 2 - bisect out of lower right vertex  \n",
    "        p2 = points[idx, 1]  \n",
    "        x = x + (1.0 / NP.sqrt(3)) * p2 * NP.cos(NP.pi / 6)\n",
    "        y = y - (1.0 / NP.sqrt(3)) * p2 * NP.sin(NP.pi / 6)        \n",
    "        # Vector 3 - bisect out of top vertex\n",
    "        p3 = points[idx, 2]\n",
    "        y = y + (1.0 / NP.sqrt(3) * p3)\n",
    "      \n",
    "        tripts[idx,:] = (x,y)\n",
    "\n",
    "    return tripts\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plotSimplex(ds, s=20);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The idea behind a `Potential` is something that is not part of the likelihood, but enforces a constraint by setting the probability to 0 if the constraint is violated. We use it here to give each cluster some membership and to order the means to remove the non-identifiability problem. See below for how its used.\n",
    "\n",
    "The sampler below has a lot of problems. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "with pm.Model() as mofb:\n",
    "    p = pm.Dirichlet('p', a=np.array([10., 10., 10.]), shape=3)\n",
    "    # ensure all clusters have some points\n",
    "    p_min_potential = pm.Potential('p_min_potential', tt.switch(tt.min(p) < .1, -np.inf, 0))\n",
    "    # cluster centers\n",
    "    means = pm.Normal('means', mu=0, sd=10, shape=3, transform=tr.ordered,\n",
    "                  testval=np.array([-1, 0, 1]))\n",
    "\n",
    "    category = pm.Categorical('category',\n",
    "                              p=p,\n",
    "                              shape=data.shape[0])\n",
    "\n",
    "    # likelihood for each observed value\n",
    "    points = pm.Normal('obs',\n",
    "                       mu=means[category],\n",
    "                       sd=1., #sds[category],\n",
    "                       observed=data)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n",
      "CompoundStep\n",
      ">NUTS: [means, p]\n",
      ">CategoricalGibbsMetropolis: [category]\n",
      "Sampling 2 chains: 100%|██████████| 21000/21000 [06:13<00:00, 56.23draws/s]\n",
      "There were 10 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "There were 7 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "The number of effective samples is smaller than 10% for some parameters.\n"
     ]
    }
   ],
   "source": [
    "with mofb:\n",
    "    tripletrace_fullb = pm.sample(10000, nuts_kwargs=dict(target_accept=0.95))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "//anaconda/envs/py3l/lib/python3.6/site-packages/matplotlib/axes/_base.py:3604: MatplotlibDeprecationWarning: \n",
      "The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
      "  alternative='`bottom`', obj_type='argument')\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trace_mofb = tripletrace_fullb[3000::5]\n",
    "pm.traceplot(trace_mofb, varnames=[\"means\", \"p\"], combined=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "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>mc_error</th>\n",
       "      <th>hpd_2.5</th>\n",
       "      <th>hpd_97.5</th>\n",
       "      <th>n_eff</th>\n",
       "      <th>Rhat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>category__0</th>\n",
       "      <td>1.406071</td>\n",
       "      <td>0.575480</td>\n",
       "      <td>0.014954</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1362.583316</td>\n",
       "      <td>1.000115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__1</th>\n",
       "      <td>0.228571</td>\n",
       "      <td>0.421610</td>\n",
       "      <td>0.010938</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1306.422985</td>\n",
       "      <td>1.000472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__2</th>\n",
       "      <td>0.395357</td>\n",
       "      <td>0.496897</td>\n",
       "      <td>0.011826</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1674.615993</td>\n",
       "      <td>1.000276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__3</th>\n",
       "      <td>0.108929</td>\n",
       "      <td>0.311550</td>\n",
       "      <td>0.008695</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1307.190266</td>\n",
       "      <td>1.001074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__4</th>\n",
       "      <td>0.824643</td>\n",
       "      <td>0.552559</td>\n",
       "      <td>0.011552</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2328.351050</td>\n",
       "      <td>0.999864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__5</th>\n",
       "      <td>0.280714</td>\n",
       "      <td>0.452516</td>\n",
       "      <td>0.011192</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1662.033357</td>\n",
       "      <td>1.000363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__6</th>\n",
       "      <td>1.948929</td>\n",
       "      <td>0.221760</td>\n",
       "      <td>0.004981</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1702.198664</td>\n",
       "      <td>1.002138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__7</th>\n",
       "      <td>1.712143</td>\n",
       "      <td>0.473583</td>\n",
       "      <td>0.012564</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1256.089935</td>\n",
       "      <td>0.999754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__8</th>\n",
       "      <td>0.900357</td>\n",
       "      <td>0.569837</td>\n",
       "      <td>0.011499</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2245.021084</td>\n",
       "      <td>0.999675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__9</th>\n",
       "      <td>1.739286</td>\n",
       "      <td>0.454218</td>\n",
       "      <td>0.012043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1421.562668</td>\n",
       "      <td>1.000357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__10</th>\n",
       "      <td>0.432500</td>\n",
       "      <td>0.506121</td>\n",
       "      <td>0.013448</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1749.523067</td>\n",
       "      <td>0.999755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__11</th>\n",
       "      <td>1.758214</td>\n",
       "      <td>0.445337</td>\n",
       "      <td>0.011746</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1291.113963</td>\n",
       "      <td>1.000946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__12</th>\n",
       "      <td>1.916071</td>\n",
       "      <td>0.282386</td>\n",
       "      <td>0.006534</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1765.304573</td>\n",
       "      <td>1.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__13</th>\n",
       "      <td>1.866429</td>\n",
       "      <td>0.342284</td>\n",
       "      <td>0.009551</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1477.039629</td>\n",
       "      <td>0.999995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__14</th>\n",
       "      <td>0.225000</td>\n",
       "      <td>0.419289</td>\n",
       "      <td>0.010990</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1472.910821</td>\n",
       "      <td>1.001048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__15</th>\n",
       "      <td>0.459286</td>\n",
       "      <td>0.513864</td>\n",
       "      <td>0.011694</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1619.209679</td>\n",
       "      <td>1.000851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__16</th>\n",
       "      <td>0.649643</td>\n",
       "      <td>0.528238</td>\n",
       "      <td>0.011004</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2333.805817</td>\n",
       "      <td>0.999647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__17</th>\n",
       "      <td>0.440000</td>\n",
       "      <td>0.511273</td>\n",
       "      <td>0.012340</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1803.682268</td>\n",
       "      <td>1.000142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__18</th>\n",
       "      <td>1.989286</td>\n",
       "      <td>0.102954</td>\n",
       "      <td>0.002396</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2055.892030</td>\n",
       "      <td>0.999691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__19</th>\n",
       "      <td>1.957143</td>\n",
       "      <td>0.202535</td>\n",
       "      <td>0.004574</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1996.770513</td>\n",
       "      <td>1.001435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__20</th>\n",
       "      <td>0.445714</td>\n",
       "      <td>0.510514</td>\n",
       "      <td>0.013029</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1827.687813</td>\n",
       "      <td>0.999651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__21</th>\n",
       "      <td>1.989643</td>\n",
       "      <td>0.101242</td>\n",
       "      <td>0.002039</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2505.025728</td>\n",
       "      <td>0.999954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__22</th>\n",
       "      <td>1.930000</td>\n",
       "      <td>0.255147</td>\n",
       "      <td>0.006387</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1414.514214</td>\n",
       "      <td>1.000427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__23</th>\n",
       "      <td>1.332143</td>\n",
       "      <td>0.587704</td>\n",
       "      <td>0.014147</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1797.023673</td>\n",
       "      <td>1.001351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__24</th>\n",
       "      <td>1.733571</td>\n",
       "      <td>0.457963</td>\n",
       "      <td>0.012479</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1184.444494</td>\n",
       "      <td>1.000431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__25</th>\n",
       "      <td>1.515714</td>\n",
       "      <td>0.556555</td>\n",
       "      <td>0.015364</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1331.826974</td>\n",
       "      <td>0.999807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__26</th>\n",
       "      <td>0.425000</td>\n",
       "      <td>0.500089</td>\n",
       "      <td>0.012622</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1455.116136</td>\n",
       "      <td>0.999694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__27</th>\n",
       "      <td>1.979643</td>\n",
       "      <td>0.141219</td>\n",
       "      <td>0.002913</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2378.339776</td>\n",
       "      <td>1.003031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__28</th>\n",
       "      <td>1.722143</td>\n",
       "      <td>0.468214</td>\n",
       "      <td>0.012763</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1469.880006</td>\n",
       "      <td>1.000669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__29</th>\n",
       "      <td>0.873214</td>\n",
       "      <td>0.554202</td>\n",
       "      <td>0.011719</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2572.986079</td>\n",
       "      <td>1.000211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__76</th>\n",
       "      <td>0.473929</td>\n",
       "      <td>0.516892</td>\n",
       "      <td>0.012308</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1568.551705</td>\n",
       "      <td>1.000044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__77</th>\n",
       "      <td>1.940714</td>\n",
       "      <td>0.237666</td>\n",
       "      <td>0.005468</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2057.523309</td>\n",
       "      <td>1.000546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__78</th>\n",
       "      <td>1.552857</td>\n",
       "      <td>0.545821</td>\n",
       "      <td>0.014423</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1278.184527</td>\n",
       "      <td>1.001185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__79</th>\n",
       "      <td>1.197143</td>\n",
       "      <td>0.603317</td>\n",
       "      <td>0.012732</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2419.356924</td>\n",
       "      <td>0.999812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__80</th>\n",
       "      <td>0.119643</td>\n",
       "      <td>0.325642</td>\n",
       "      <td>0.008502</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1457.209338</td>\n",
       "      <td>0.999644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__81</th>\n",
       "      <td>0.671429</td>\n",
       "      <td>0.535762</td>\n",
       "      <td>0.011043</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2416.678137</td>\n",
       "      <td>0.999991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__82</th>\n",
       "      <td>1.471786</td>\n",
       "      <td>0.560539</td>\n",
       "      <td>0.014711</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1408.654254</td>\n",
       "      <td>0.999643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__83</th>\n",
       "      <td>0.687857</td>\n",
       "      <td>0.540894</td>\n",
       "      <td>0.011061</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2184.065312</td>\n",
       "      <td>1.000089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__84</th>\n",
       "      <td>1.832500</td>\n",
       "      <td>0.379117</td>\n",
       "      <td>0.009391</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1442.318862</td>\n",
       "      <td>0.999651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__85</th>\n",
       "      <td>0.238929</td>\n",
       "      <td>0.428101</td>\n",
       "      <td>0.010072</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1580.222999</td>\n",
       "      <td>0.999649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__86</th>\n",
       "      <td>1.001786</td>\n",
       "      <td>0.574764</td>\n",
       "      <td>0.011643</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2431.781980</td>\n",
       "      <td>0.999884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__87</th>\n",
       "      <td>0.714286</td>\n",
       "      <td>0.546884</td>\n",
       "      <td>0.011661</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2298.065679</td>\n",
       "      <td>0.999726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__88</th>\n",
       "      <td>1.627143</td>\n",
       "      <td>0.519869</td>\n",
       "      <td>0.014501</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1153.871436</td>\n",
       "      <td>0.999673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__89</th>\n",
       "      <td>1.979286</td>\n",
       "      <td>0.142426</td>\n",
       "      <td>0.002728</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2325.559161</td>\n",
       "      <td>0.999668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__90</th>\n",
       "      <td>1.987143</td>\n",
       "      <td>0.112658</td>\n",
       "      <td>0.002501</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1955.167689</td>\n",
       "      <td>1.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__91</th>\n",
       "      <td>1.411429</td>\n",
       "      <td>0.573845</td>\n",
       "      <td>0.012387</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2121.534066</td>\n",
       "      <td>0.999742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__92</th>\n",
       "      <td>0.287857</td>\n",
       "      <td>0.454339</td>\n",
       "      <td>0.011087</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1656.472820</td>\n",
       "      <td>1.000951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__93</th>\n",
       "      <td>0.120714</td>\n",
       "      <td>0.325795</td>\n",
       "      <td>0.009032</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1414.023539</td>\n",
       "      <td>0.999648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__94</th>\n",
       "      <td>0.817857</td>\n",
       "      <td>0.560327</td>\n",
       "      <td>0.011872</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1951.110342</td>\n",
       "      <td>0.999649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__95</th>\n",
       "      <td>1.003571</td>\n",
       "      <td>0.596407</td>\n",
       "      <td>0.012872</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2002.102565</td>\n",
       "      <td>0.999679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__96</th>\n",
       "      <td>0.076786</td>\n",
       "      <td>0.266251</td>\n",
       "      <td>0.007683</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1365.135665</td>\n",
       "      <td>0.999645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__97</th>\n",
       "      <td>1.140000</td>\n",
       "      <td>0.584661</td>\n",
       "      <td>0.010782</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2331.457873</td>\n",
       "      <td>0.999697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__98</th>\n",
       "      <td>1.993571</td>\n",
       "      <td>0.079920</td>\n",
       "      <td>0.001462</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2601.475216</td>\n",
       "      <td>1.000921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category__99</th>\n",
       "      <td>0.199643</td>\n",
       "      <td>0.400624</td>\n",
       "      <td>0.010216</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1420.458431</td>\n",
       "      <td>1.000063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>p__0</th>\n",
       "      <td>0.360226</td>\n",
       "      <td>0.081588</td>\n",
       "      <td>0.003317</td>\n",
       "      <td>0.204306</td>\n",
       "      <td>0.519089</td>\n",
       "      <td>519.487415</td>\n",
       "      <td>1.002060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>p__1</th>\n",
       "      <td>0.325225</td>\n",
       "      <td>0.071761</td>\n",
       "      <td>0.001919</td>\n",
       "      <td>0.172866</td>\n",
       "      <td>0.459570</td>\n",
       "      <td>1294.128430</td>\n",
       "      <td>0.999978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>p__2</th>\n",
       "      <td>0.314550</td>\n",
       "      <td>0.065550</td>\n",
       "      <td>0.002548</td>\n",
       "      <td>0.189302</td>\n",
       "      <td>0.444908</td>\n",
       "      <td>587.062612</td>\n",
       "      <td>1.001335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>means__0</th>\n",
       "      <td>-1.886959</td>\n",
       "      <td>0.263363</td>\n",
       "      <td>0.007660</td>\n",
       "      <td>-2.388613</td>\n",
       "      <td>-1.365366</td>\n",
       "      <td>1295.486095</td>\n",
       "      <td>1.002200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>means__1</th>\n",
       "      <td>-0.407361</td>\n",
       "      <td>0.565056</td>\n",
       "      <td>0.029250</td>\n",
       "      <td>-1.451171</td>\n",
       "      <td>0.683029</td>\n",
       "      <td>286.606094</td>\n",
       "      <td>1.002899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>means__2</th>\n",
       "      <td>1.948775</td>\n",
       "      <td>0.303411</td>\n",
       "      <td>0.009837</td>\n",
       "      <td>1.418676</td>\n",
       "      <td>2.599857</td>\n",
       "      <td>791.034984</td>\n",
       "      <td>1.001906</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>106 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  mean        sd  mc_error   hpd_2.5  hpd_97.5        n_eff      Rhat\n",
       "category__0   1.406071  0.575480  0.014954  1.000000  2.000000  1362.583316  1.000115\n",
       "category__1   0.228571  0.421610  0.010938  0.000000  1.000000  1306.422985  1.000472\n",
       "category__2   0.395357  0.496897  0.011826  0.000000  1.000000  1674.615993  1.000276\n",
       "category__3   0.108929  0.311550  0.008695  0.000000  1.000000  1307.190266  1.001074\n",
       "category__4   0.824643  0.552559  0.011552  0.000000  2.000000  2328.351050  0.999864\n",
       "category__5   0.280714  0.452516  0.011192  0.000000  1.000000  1662.033357  1.000363\n",
       "category__6   1.948929  0.221760  0.004981  1.000000  2.000000  1702.198664  1.002138\n",
       "category__7   1.712143  0.473583  0.012564  1.000000  2.000000  1256.089935  0.999754\n",
       "category__8   0.900357  0.569837  0.011499  0.000000  2.000000  2245.021084  0.999675\n",
       "category__9   1.739286  0.454218  0.012043  1.000000  2.000000  1421.562668  1.000357\n",
       "category__10  0.432500  0.506121  0.013448  0.000000  1.000000  1749.523067  0.999755\n",
       "category__11  1.758214  0.445337  0.011746  1.000000  2.000000  1291.113963  1.000946\n",
       "category__12  1.916071  0.282386  0.006534  1.000000  2.000000  1765.304573  1.000003\n",
       "category__13  1.866429  0.342284  0.009551  1.000000  2.000000  1477.039629  0.999995\n",
       "category__14  0.225000  0.419289  0.010990  0.000000  1.000000  1472.910821  1.001048\n",
       "category__15  0.459286  0.513864  0.011694  0.000000  1.000000  1619.209679  1.000851\n",
       "category__16  0.649643  0.528238  0.011004  0.000000  1.000000  2333.805817  0.999647\n",
       "category__17  0.440000  0.511273  0.012340  0.000000  1.000000  1803.682268  1.000142\n",
       "category__18  1.989286  0.102954  0.002396  2.000000  2.000000  2055.892030  0.999691\n",
       "category__19  1.957143  0.202535  0.004574  2.000000  2.000000  1996.770513  1.001435\n",
       "category__20  0.445714  0.510514  0.013029  0.000000  1.000000  1827.687813  0.999651\n",
       "category__21  1.989643  0.101242  0.002039  2.000000  2.000000  2505.025728  0.999954\n",
       "category__22  1.930000  0.255147  0.006387  1.000000  2.000000  1414.514214  1.000427\n",
       "category__23  1.332143  0.587704  0.014147  0.000000  2.000000  1797.023673  1.001351\n",
       "category__24  1.733571  0.457963  0.012479  1.000000  2.000000  1184.444494  1.000431\n",
       "category__25  1.515714  0.556555  0.015364  1.000000  2.000000  1331.826974  0.999807\n",
       "category__26  0.425000  0.500089  0.012622  0.000000  1.000000  1455.116136  0.999694\n",
       "category__27  1.979643  0.141219  0.002913  2.000000  2.000000  2378.339776  1.003031\n",
       "category__28  1.722143  0.468214  0.012763  1.000000  2.000000  1469.880006  1.000669\n",
       "category__29  0.873214  0.554202  0.011719  0.000000  2.000000  2572.986079  1.000211\n",
       "...                ...       ...       ...       ...       ...          ...       ...\n",
       "category__76  0.473929  0.516892  0.012308  0.000000  1.000000  1568.551705  1.000044\n",
       "category__77  1.940714  0.237666  0.005468  1.000000  2.000000  2057.523309  1.000546\n",
       "category__78  1.552857  0.545821  0.014423  1.000000  2.000000  1278.184527  1.001185\n",
       "category__79  1.197143  0.603317  0.012732  0.000000  2.000000  2419.356924  0.999812\n",
       "category__80  0.119643  0.325642  0.008502  0.000000  1.000000  1457.209338  0.999644\n",
       "category__81  0.671429  0.535762  0.011043  0.000000  1.000000  2416.678137  0.999991\n",
       "category__82  1.471786  0.560539  0.014711  1.000000  2.000000  1408.654254  0.999643\n",
       "category__83  0.687857  0.540894  0.011061  0.000000  1.000000  2184.065312  1.000089\n",
       "category__84  1.832500  0.379117  0.009391  1.000000  2.000000  1442.318862  0.999651\n",
       "category__85  0.238929  0.428101  0.010072  0.000000  1.000000  1580.222999  0.999649\n",
       "category__86  1.001786  0.574764  0.011643  0.000000  2.000000  2431.781980  0.999884\n",
       "category__87  0.714286  0.546884  0.011661  0.000000  1.000000  2298.065679  0.999726\n",
       "category__88  1.627143  0.519869  0.014501  1.000000  2.000000  1153.871436  0.999673\n",
       "category__89  1.979286  0.142426  0.002728  2.000000  2.000000  2325.559161  0.999668\n",
       "category__90  1.987143  0.112658  0.002501  2.000000  2.000000  1955.167689  1.000005\n",
       "category__91  1.411429  0.573845  0.012387  1.000000  2.000000  2121.534066  0.999742\n",
       "category__92  0.287857  0.454339  0.011087  0.000000  1.000000  1656.472820  1.000951\n",
       "category__93  0.120714  0.325795  0.009032  0.000000  1.000000  1414.023539  0.999648\n",
       "category__94  0.817857  0.560327  0.011872  0.000000  2.000000  1951.110342  0.999649\n",
       "category__95  1.003571  0.596407  0.012872  0.000000  2.000000  2002.102565  0.999679\n",
       "category__96  0.076786  0.266251  0.007683  0.000000  1.000000  1365.135665  0.999645\n",
       "category__97  1.140000  0.584661  0.010782  0.000000  2.000000  2331.457873  0.999697\n",
       "category__98  1.993571  0.079920  0.001462  2.000000  2.000000  2601.475216  1.000921\n",
       "category__99  0.199643  0.400624  0.010216  0.000000  1.000000  1420.458431  1.000063\n",
       "p__0          0.360226  0.081588  0.003317  0.204306  0.519089   519.487415  1.002060\n",
       "p__1          0.325225  0.071761  0.001919  0.172866  0.459570  1294.128430  0.999978\n",
       "p__2          0.314550  0.065550  0.002548  0.189302  0.444908   587.062612  1.001335\n",
       "means__0     -1.886959  0.263363  0.007660 -2.388613 -1.365366  1295.486095  1.002200\n",
       "means__1     -0.407361  0.565056  0.029250 -1.451171  0.683029   286.606094  1.002899\n",
       "means__2      1.948775  0.303411  0.009837  1.418676  2.599857   791.034984  1.001906\n",
       "\n",
       "[106 rows x 7 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.summary(trace_mofb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making Problems go away\n",
    "\n",
    "A lot will go away when identifiability improves through separated gaussians. But that changes the data. If we want any further improvement on this data, we are going to have to stop sampling so many discrete categoricals. And for that we will need a marginalization trick."
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}