Talk.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2 style=\"display:inline;\"> Turning </h2> \n",
    "<img src=\"img/pymc3.jpg\" alt=\"pymc3_logo\" style=\"width: 200px;display:inline;\"/>\n",
    "<h2 style=\"display:inline;\"> into </h2>\n",
    "<img src=\"img/sklearn.png\" alt=\"sklearn_logo\" style=\"width: 150px;display:inline;\"/>\n",
    "\n",
    "<h3>\n",
    "Nicole Carlson <br>\n",
    "@parsing_science <br>\n",
    "github.com/parsing-science\n",
    "</h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Overview </h2>\n",
    "- Motivating Example: Scikitlearn Linear Regression\n",
    "- Intro to PyMC3\n",
    "- PyMC3 Linear Regression\n",
    "- Compare Scikitlearn and PyMC3\n",
    "- Implement Hiearchical Logistic Regression in PyMC3\n",
    "- Use the Hierarchical Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "from pymc3 import summary, traceplot\n",
    "import pymc3 as pm\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import r2_score\n",
    "\n",
    "import theano.tensor as T\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(action=\"ignore\", module=\"scipy\", message=\"^internal gelsd\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "plt.rcParams['figure.figsize'] = (15, 10)\n",
    "\n",
    "plt.rcParams['font.size'] = 16"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2>Motivating Example: Scikitlearn</h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Generate Fake Data </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "X = np.random.randn(1000, 1)\n",
    "noise = 2 * np.random.randn(1000, 1)\n",
    "Y = 4 * X + 3 + noise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x10983acd0>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA38AAAJHCAYAAAAg6e/7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nuwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10978e710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Create a LinearRegression instance and use it </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "LR = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LR.fit(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "X_predict = LR.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.80613292900134936"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LR.score(X, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Save and Load LR </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "pickle.dump(LR, open('LR.pickle', 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "saved_LR = pickle.load(open('LR.pickle', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.80613292900134936"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "saved_LR.score(X, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Intro to PyMC3 </h2>\n",
    "\n",
    "- Library for Probabilistic Modeling\n",
    "- MCMC Samplers\n",
    "- Variational Fitting Algorithms\n",
    "- Built on top of Theano\n",
    "- Completely written in Python"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Using PyMC3 </h2>\n",
    "- Step 1: Set up: Parameterize your model, choose priors, and insert training data\n",
    "- Step 2: Inference: infer your parameters using sampling (e.g. NUTS) or variational inference (e.g. ADVI)\n",
    "- Step 3: Interpret: Check your parameter distributions and model fit\n",
    "- Step 4: Predict data: Create posterior samples with your inferred parameters"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 1: Set up the PyMC3 model </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "lin_reg_model = pm.Model()\n",
    "with lin_reg_model:\n",
    "\n",
    "    alpha = pm.Normal('alpha', mu=0, tau=10.**-2, shape=(1))\n",
    "    betas = pm.Normal('betas', mu=0, tau=10. ** -2, shape=(1))\n",
    "    \n",
    "    s = pm.HalfNormal('s', tau=1)\n",
    "    \n",
    "    temp = alpha + betas * X\n",
    "\n",
    "    y = pm.Normal('y', mu=temp , tau=s ** -2, observed=Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 2: Infer your parameters </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2000/2000.0 [00:03<00:00, 572.71it/s]\n"
     ]
    }
   ],
   "source": [
    "with lin_reg_model:\n",
    "    step = pm.NUTS()\n",
    "    nuts_trace = pm.sample(2e3, step)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 3: Interpret your parameters </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "alpha:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  2.993            0.280            0.022            [2.886, 3.143]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  2.879          2.971          3.015          3.055          3.140\n",
      "\n",
      "\n",
      "betas:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  3.973            0.390            0.032            [3.873, 4.123]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  3.877          3.960          4.005          4.048          4.129\n",
      "\n",
      "\n",
      "s:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  2.021            0.698            0.045            [1.876, 2.073]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  1.888          1.940          1.972          2.001          2.095\n",
      "\n"
     ]
    }
   ],
   "source": [
    "summary(nuts_trace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x10e60b050>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x10e66ec10>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x10e6e5390>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x10e752e50>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x10e76de50>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x10e859090>]], dtype=object)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0oAAAGbCAYAAAD6J2BOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nc85V+X/5onlHhjeOYQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10e5ce590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "traceplot(nuts_trace)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2>Step 4: Predict data by creating posterior predictive samples </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [01:57<00:00, 10.66it/s]\n"
     ]
    }
   ],
   "source": [
    "ppc = pm.sample_ppc(\n",
    "    nuts_trace[1000:],\n",
    "    model=lin_reg_model,\n",
    "    samples=1000\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "pred = ppc['y'].mean(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.80553921760805347"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2_score(Y, pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Most PyMC3 Examples </h2>\n",
    "\n",
    "* Exist in Jupyter notebooks\n",
    "* Focus on training a model with one dataset\n",
    "* Don’t demonstrate how to predict new data\n",
    "* Don’t show how to save your model for later use\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Comparing the two libraries </h2>\n",
    "\n",
    "<table>\n",
    "<tr>\n",
    "<th> </th>\n",
    "<th>sklearn</th>\n",
    "<th>PyMC3</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Find model parameters</th>\n",
    "<th>Easy</th>\n",
    "<th>Medium</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Predict new data</th>\n",
    "<th>Easy</th>\n",
    "<th>Difficult</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Score a model</th>\n",
    "<th>Easy</th>\n",
    "<th>Difficult</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Save a trained model</th>\n",
    "<th>Easy</th>\n",
    "<th>Impossible?</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Load a trained model</th>\n",
    "<th>Easy</th>\n",
    "<th>Impossible?</th>\n",
    "</tr>\n",
    "</table>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> What we want to reproduce in PyMC3 </h2>\n",
    "\n",
    "* fit\n",
    "* predict\n",
    "* score\n",
    "* save\n",
    "* load\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Map between Scikitlearn and PyMC3 </h2>\n",
    "\n",
    "<table>\n",
    "<tr>\n",
    "<th>sklearn</th>\n",
    "<th>PyMC3</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Fit</th>\n",
    "<th>Step 1: Set up, Step 2: Inference</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Predict</th>\n",
    "<th>Step 4: Predict data</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Score</th>\n",
    "<th>Step 4: Predict data</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>Save/Load</th>\n",
    "<th>??</th>\n",
    "</tr>\n",
    "<tr>\n",
    "<th>??</th>\n",
    "<th>Step 3: Interpret</th>\n",
    "</tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Initializing the HLM class </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "class HLM(BaseEstimator):\n",
    "    \"\"\"\n",
    "    Custom Hierachical Linear Model built using PyMC3.\n",
    "    \"\"\"\n",
    "\n",
    "    def __init__(self):\n",
    "        self.cached_model = None\n",
    "        self.shared_vars = None\n",
    "        self.num_cats = None\n",
    "        self.num_pred = None\n",
    "        self.advi_trace = None\n",
    "        self.v_params = None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Fit </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def fit(self, X, cats, y):\n",
    "    \"\"\"\n",
    "    Train the HLM model\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    X : numpy array, shape [n_samples, n_features]\n",
    "\n",
    "    cats: numpy array, shape [n_samples, ]\n",
    "\n",
    "    y : numpy array, shape [n_samples, ]\n",
    "    \"\"\"\n",
    "    self.num_cats = len(np.unique(cats))\n",
    "    num_samples, self.num_pred = X.shape\n",
    "\n",
    "    if self.cached_model is None:\n",
    "        # Step 1: Set up the model\n",
    "        self.cached_model, o = self.create_model()\n",
    "\n",
    "    #Step 2: Inference\n",
    "    minibatch_tensors = [\n",
    "        self.shared_vars['model_input'],\n",
    "        self.shared_vars['model_output'],\n",
    "        self.shared_vars['model_cats']\n",
    "    ]\n",
    "    minibatch_RVs = [o]\n",
    "\n",
    "    minibatches = self._create_minibatch([X, y, cats], num_samples)\n",
    "\n",
    "    self.v_params, self.advi_trace = self._inference(\n",
    "        minibatch_tensors,\n",
    "        minibatch_RVs,\n",
    "        minibatches,\n",
    "        num_samples\n",
    "    )\n",
    "\n",
    "    return self"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 1: Set up the model </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def create_model(self):\n",
    "    \"\"\"\n",
    "    Creates and returns the PyMC3 model.\n",
    "\n",
    "    Returns the model and the output variable. The latter is for use in ADVI minibatch.\n",
    "    \"\"\"\n",
    "    model_input = theano.shared(np.zeros([1, self.num_pred]))\n",
    "\n",
    "    model_output = theano.shared(np.zeros(1))\n",
    "\n",
    "    model_cats = theano.shared(np.zeros(1, dtype='int'))\n",
    "\n",
    "    self.shared_vars = {\n",
    "        'model_input': model_input,\n",
    "        'model_output': model_output,\n",
    "        'model_cats': model_cats\n",
    "    }\n",
    "\n",
    "    model = pm.Model()\n",
    "\n",
    "    with model:\n",
    "        # Both alpha and beta are drawn from Normal distributions\n",
    "        mu_alpha = pm.Normal(\"mu_alpha\", mu=0, sd=10)\n",
    "        sigma_alpha = pm.HalfNormal(\"sigma_alpha\", sd=10)\n",
    "\n",
    "        mu_beta = pm.Normal(\"mu_beta\", mu=0, sd=10)\n",
    "        sigma_beta = pm.HalfNormal(\"sigma_beta\", sd=10)\n",
    "\n",
    "        alpha = pm.Normal(\n",
    "            'alpha',\n",
    "            mu=mu_alpha,\n",
    "            sd=sigma_alpha,\n",
    "            shape=(self.num_cats,)\n",
    "        )\n",
    "\n",
    "        beta = pm.Normal(\n",
    "            'beta',\n",
    "            mu=mu_beta,\n",
    "            sd=sigma_beta,\n",
    "            shape=(self.num_cats, self.num_pred)\n",
    "        )\n",
    "\n",
    "        c = model_cats\n",
    "\n",
    "        temp = alpha[c] + T.sum(beta[c] * model_input, 1)\n",
    "\n",
    "        p = pm.invlogit(temp)\n",
    "\n",
    "        o = pm.Bernoulli('o', p, observed=model_output)\n",
    "\n",
    "    return model, o"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 2a: Inference: Create minibatches for ADVI</h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "@staticmethod\n",
    "def _create_minibatch(data, num_samples, size=100):\n",
    "    \"\"\"\n",
    "    Generator that returns mini-batches in each iteration\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        # Return random data samples of set size each iteration\n",
    "        ixs = np.random.randint(num_samples, size=size)\n",
    "        yield [tensor[ixs] for tensor in data]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Step 2: Inference </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def _inference(\n",
    "    self,\n",
    "    minibatch_tensors,\n",
    "    minibatch_RVs,\n",
    "    minibatches,\n",
    "    num_samples\n",
    "):\n",
    "    \"\"\"\n",
    "    Runs minibatch variational ADVI and then sample from those results.\n",
    "    \"\"\"\n",
    "    with self.cached_model:\n",
    "        v_params = pm.variational.advi_minibatch(\n",
    "            n=50000,\n",
    "            minibatch_tensors=minibatch_tensors,\n",
    "            minibatch_RVs=minibatch_RVs,\n",
    "            minibatches=minibatches,\n",
    "            total_size=int(num_samples),\n",
    "            learning_rate=1e-2,\n",
    "            epsilon=1.0\n",
    "        )\n",
    "\n",
    "        advi_trace = pm.variational.sample_vp(v_params, draws=7500)\n",
    "\n",
    "    return v_params, advi_trace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Predict Probabilities </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def predict_proba(self, X, cats):\n",
    "    \"\"\"\n",
    "    Predicts probabilities of new data with a trained HLM\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    X : numpy array, shape [n_samples, n_features]\n",
    "\n",
    "    cats: numpy array, shape [n_samples, ]\n",
    "    \"\"\"\n",
    "\n",
    "    if self.advi_trace is None:\n",
    "        raise PSToolkitError(\"Run fit on the model before predict.\")\n",
    "\n",
    "    num_samples = X.shape[0]\n",
    "\n",
    "    if self.cached_model is None:\n",
    "        self.cached_model, o = self.create_model()\n",
    "    \n",
    "    # Step 4: Predict data\n",
    "    self._set_shared_vars(X, np.zeros(num_samples), cats)\n",
    "   \n",
    "    ppc = pm.sample_ppc(\n",
    "        self.advi_trace,\n",
    "        model=self.cached_model,\n",
    "        samples=2000\n",
    "    )\n",
    "\n",
    "    return ppc['o'].mean(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Set Shared Variables (part of Step 4) </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def _set_shared_vars(self, model_input, model_output, model_cats):\n",
    "    \"\"\"\n",
    "    Creates theano shared variables for the PyMC3 model.\n",
    "    \"\"\"\n",
    "    self.shared_vars['model_input'].set_value(model_input)\n",
    "\n",
    "    self.shared_vars['model_output'].set_value(model_output)\n",
    "\n",
    "    self.shared_vars['model_cats'].set_value(model_cats)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Predict </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def predict(self, X, cats):\n",
    "    \"\"\"\n",
    "    Predicts labels of new data with a trained HLM\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    X : numpy array, shape [n_samples, n_features]\n",
    "\n",
    "    cats: numpy array, shape [n_samples, ]\n",
    "    \"\"\"\n",
    "    ppc_mean = self.predict_proba(X, cats)\n",
    "\n",
    "    pred = ppc_mean > 0.5\n",
    "\n",
    "    return pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Score </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def score(self, X, cats, y):\n",
    "    \"\"\"\n",
    "    Scores new data with a trained model.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    X : numpy array, shape [n_samples, n_features]\n",
    "\n",
    "    cats: numpy array, shape [n_samples, ]\n",
    "\n",
    "    y : numpy array, shape [n_samples, ]\n",
    "    \"\"\"\n",
    "    return accuracy_score(y, self.predict(X, cats))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Save </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def save(self, file_prefix):\n",
    "    \"\"\"\n",
    "    Saves the advi_trace, v_params, and param files with the given file_prefix.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    file_prefix: str\n",
    "    path and prefix used to identify where to save trace and params for this model.\n",
    "    ex. given file_prefix = \"path/to/file/\"\n",
    "    This will attempt to save to \"path/to/file/advi_trace.pickle\" and \"path/to/file/params.pickle\"\n",
    "    \"\"\"\n",
    "    fileObject = open(file_prefix + \"advi_trace.pickle\", 'w')\n",
    "    joblib.dump(self.advi_trace, fileObject)\n",
    "    fileObject.close()\n",
    "\n",
    "    fileObject = open(file_prefix + \"v_params.pickle\", 'w')\n",
    "    joblib.dump(self.v_params, fileObject)\n",
    "    fileObject.close()\n",
    "\n",
    "    fileObject = open(file_prefix + \"params.pickle\", 'w')\n",
    "    joblib.dump(\n",
    "        {\"num_cats\": self.num_cats, \"num_pred\": self.num_pred},\n",
    "        fileObject\n",
    "    )\n",
    "    fileObject.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Load </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def load(self, file_prefix):\n",
    "    \"\"\"\n",
    "    Loads a saved version of the advi_trace, v_params, and param files with the given file_prefix.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    file_prefix: str\n",
    "    path and prefix used to identify where to load saved trace and params for this model.\n",
    "    ex. given file_prefix = \"path/to/file/\"\n",
    "    This will attempt to load \"path/to/file/advi_trace.pickle\" and \"path/to/file/params.pickle\"\n",
    "    \"\"\"\n",
    "    self.advi_trace = joblib.load(file_prefix + \"advi_trace.pickle\")\n",
    "    self.v_params = joblib.load(file_prefix + \"v_params.pickle\")\n",
    "\n",
    "    params = joblib.load(file_prefix + \"params.pickle\")\n",
    "    self.num_cats = params[\"num_cats\"]\n",
    "    self.num_pred = params[\"num_pred\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2>Use the Custom PyMC3 Class </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Generate Fake Data </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "num_pred = 2\n",
    "num_samples = 50000\n",
    "num_categories = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "alphas = 5 * np.random.randn(num_categories) + 5 # mu_alpha = sigma_alpha = 5\n",
    "betas = 10 * np.random.randn(num_categories, num_pred) + 10 # mu_beta = sigma_beta = 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.15856219,  4.02812109])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alphas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  7.49863681,   9.0573482 ],\n",
       "       [ -2.53113336,  23.23233084]])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "betas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "def numpy_invlogit(x):\n",
    "    return 1 / (1 + np.exp(-x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "x_a = np.random.randn(num_samples, num_pred)\n",
    "y_a = np.random.binomial(1, numpy_invlogit(alphas[0] + np.sum(betas[0] * x_a, 1)))\n",
    "x_b = np.random.randn(num_samples, num_pred)\n",
    "y_b = np.random.binomial(1, numpy_invlogit(alphas[1] + np.sum(betas[1] * x_b, 1)))\n",
    "\n",
    "X = np.concatenate([x_a, x_b])\n",
    "Y = np.concatenate([y_a, y_b])\n",
    "cats = np.concatenate([\n",
    "    np.zeros(num_samples, dtype=np.int),\n",
    "    np.ones(num_samples, dtype=np.int)\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Load the class </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "from ps_toolkit import HLM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "HLM1 = HLM()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "warnings.filterwarnings(action=\"ignore\", module=\"theano\", message=\"^theano.tensor.round\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Fit the HLM </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Average ELBO = -8,122.32: 100%|██████████| 50000/50000 [00:24<00:00, 2031.99it/s]\n",
      "Finished minibatch ADVI: ELBO = -10,423.93\n",
      "100%|██████████| 7500/7500 [00:01<00:00, 7320.77it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "HLM()"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HLM1.fit(X, cats, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Predict and score the model </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2000/2000 [00:32<00:00, 61.88it/s]\n"
     ]
    }
   ],
   "source": [
    "probs = HLM1.predict_proba(X, cats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2000/2000 [00:30<00:00, 66.15it/s]\n"
     ]
    }
   ],
   "source": [
    "pred = HLM1.predict(X, cats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2000/2000 [00:28<00:00, 70.09it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.96577999999999997"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HLM1.score(X, cats, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Save and Load the model </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "HLM1.save(\"HLM_jar/\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "HLM2 = HLM()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "HLM2.load(\"HLM_jar/\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2000/2000 [00:33<00:00, 60.23it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.96589999999999998"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HLM2.score(X, cats, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2>Step 3: Interpret your parameters</h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAggAAAFXCAYAAADHzLbcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nA1SaMIqjkNFrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x14e65c550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "HLM1.plot_elbo()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "mu_alpha:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  2.707            2.681            0.033            [-2.503, 7.875]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  -2.479         0.924          2.682          4.527          7.913\n",
      "\n",
      "\n",
      "mu_beta:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  7.489            4.976            0.057            [-2.638, 16.969]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  -2.346         4.162          7.456          10.849         17.359\n",
      "\n",
      "\n",
      "alpha:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  1.199            0.027            0.000            [1.147, 1.251]\n",
      "  4.200            0.051            0.001            [4.102, 4.301]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  1.146          1.181          1.199          1.217          1.251\n",
      "  4.100          4.165          4.201          4.235          4.300\n",
      "\n",
      "\n",
      "beta:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  ..............................[0, :]...............................\n",
      "  7.704            0.029            0.000            [7.649, 7.762]\n",
      "  9.230            0.038            0.000            [9.158, 9.304]\n",
      "  ..............................[1, :]...............................\n",
      "  -2.530           0.037            0.000            [-2.601, -2.457]\n",
      "  24.189           0.264            0.003            [23.675, 24.699]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  .............................[0, :].............................\n",
      "  7.647          7.685          7.704          7.724          7.761\n",
      "  9.157          9.205          9.230          9.256          9.303\n",
      "  .............................[1, :].............................\n",
      "  -2.604         -2.555         -2.530         -2.505         -2.458\n",
      "  23.668         24.014         24.191         24.368         24.694\n",
      "\n",
      "\n",
      "sigma_alpha:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  5.504            3.554            0.037            [0.916, 12.275]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  1.429          3.129          4.595          6.851          14.727\n",
      "\n",
      "\n",
      "sigma_beta:\n",
      "\n",
      "  Mean             SD               MC Error         95% HPD interval\n",
      "  -------------------------------------------------------------------\n",
      "  \n",
      "  11.615           3.869            0.043            [5.141, 19.352]\n",
      "\n",
      "  Posterior quantiles:\n",
      "  2.5            25             50             75             97.5\n",
      "  |--------------|==============|==============|--------------|\n",
      "  \n",
      "  5.837          8.900          11.023         13.642         20.937\n",
      "\n"
     ]
    }
   ],
   "source": [
    "summary(HLM1.advi_trace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.15856219,  4.02812109])"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alphas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  7.49863681,   9.0573482 ],\n",
       "       [ -2.53113336,  23.23233084]])"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "betas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x103410050>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13e3da610>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x13e8280d0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13e87d490>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x13e8f2090>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13e88fad0>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x13e9cb490>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13eff68d0>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x13f065b10>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13f0ecb50>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x13f15f410>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x13f1e56d0>]], dtype=object)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1QAAANUCAYAAAC0YWxdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10341d7d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "traceplot(HLM1.advi_trace)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Next Steps </h2>\n",
    "* Build Bayesian Model base class \n",
    "* Pass more options to the init function\n",
    "* Test if this class can be used in an sklearn pipeline\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<h2> Thanks: </h2>\n",
    "- Daniel Little\n",
    "- Tudor Radoaca\n",
    "- Nico Sandller\n",
    "- PyData Chicago\n",
    "\n",
    "<h3>\n",
    "@parsing_science <br>\n",
    "github.com/parsing-science\n",
    "</h3>"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}