not_mnist.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install master branches of theano and lasagne\n",
    "[Here are the instructions](http://lasagne.readthedocs.io/en/latest/user/installation.html#bleeding-edge-version)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "sns.set_style('white')\n",
    "sns.set_context('talk')\n",
    "\n",
    "import pymc3 as pm\n",
    "import theano.tensor as tt\n",
    "import theano\n",
    "\n",
    "from scipy.stats import mode, chisquare\n",
    "from sklearn.metrics import confusion_matrix, accuracy_score\n",
    "import lasagne\n",
    "import os\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(u'1.0.1', '0.2.dev1')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#(u'1.0.1', '0.2.dev1')\n",
    "theano.__version__, lasagne.__version__\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### notMNIST images courtesy Y. Bulatov\n",
    "[notMNIST](http://yaroslavvb.blogspot.com/2011/09/notmnist-dataset.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Extract tar file\n",
    "basedir = 'data/notMNIST/'\n",
    "localfile = os.path.join(basedir, 'notMNIST_small.tar.gz')\n",
    "import tarfile\n",
    "tar = tarfile.open(localfile)\n",
    "tar.extractall()\n",
    "tar.close()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocess notMNIST\n",
    "Adapted from code by Y. Bulatov \n",
    "Source: http://yaroslavvb.com/upload/notMNIST/matlab_convert.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/notMNIST/notMNIST_small/\n",
      "Found 18726 files\n",
      "exception 'module' object has no attribute 'register_decoder'\n",
      "exception 'module' object has no attribute 'register_decoder'\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACYgAAAcXCAYAAABuYAfPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112a1e750>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAEkCAYAAADghueSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x113c190d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!/usr/bin/python2.6\n",
    "# Turns directory of notMNIST glyphs into train, test datasets\n",
    "#\n",
    "# dataset_dir --> A --> img1.png\n",
    "#                   --> img2.png\n",
    "#             --> B --> img1.png\n",
    "#                   --> img2.png\n",
    "#      ...\n",
    "#             --> J --> img1.png\n",
    "#                   --> img2.png\n",
    "#\n",
    "# Show some images with their labels\n",
    "#   load('target.mat')\n",
    "#   for i=1:5\n",
    "#     figure('Name',num2str(labels(i))),imshow(images(:,:,i)/255)\n",
    "#   end\n",
    "#\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "from scipy.io import savemat\n",
    "import numpy as np, glob, sys, os\n",
    "import PIL.Image as Image\n",
    "from matplotlib.pyplot import imshow\n",
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "def reshape_X(X_pre):\n",
    "    X_post = np.zeros((X_pre.shape[0], 1, 28, 28))\n",
    "    for i in range(X_pre.shape[0]):\n",
    "        X_post[i, 0, :, :] = X_pre[i, :, :]\n",
    "    return X_post\n",
    "\n",
    "mapping = {}\n",
    "def generate_dataset(folder):\n",
    "    print folder\n",
    "    max_count = 0\n",
    "    for (root, dirs, files) in os.walk(folder):\n",
    "        for f in files:\n",
    "            if f.endswith('.png'):\n",
    "                max_count+=1\n",
    "\n",
    "    print 'Found %s files'%(max_count,)\n",
    "    data = np.zeros((28,28,max_count))\n",
    "    labels = np.zeros((max_count,))\n",
    "    count = 0\n",
    "    for (root, dirs, files) in os.walk(folder):\n",
    "        for f in files:\n",
    "            if f.endswith('.png'):\n",
    "                try:\n",
    "                    #print(\"opening {}, {}, {}\".format(root, f, os.path.join(root, f)))\n",
    "                    img = Image.open(os.path.join(root, f));\n",
    "                    #print(\"{}, {}\".format(f, dirs))\n",
    "                    data[:,:,count]=np.asarray(img)\n",
    "                    surround_folder = os.path.split(root)[-1]\n",
    "                    assert len(surround_folder)==1\n",
    "                    ordinal = ord(surround_folder)-ord('A')\n",
    "                    labels[count]=ordinal\n",
    "                    mapping[ordinal] = surround_folder\n",
    "                    count+=1\n",
    "                except Exception as e:\n",
    "                    print(\"exception {}\".format(e))\n",
    "                    pass\n",
    "    \n",
    "    # Plot NUM samples\n",
    "    fig = plt.figure()\n",
    "    COLS, NUM = 3, 8\n",
    "    np.random.seed(NUM)\n",
    "    for i in range(NUM):\n",
    "        ax = fig.add_subplot(COLS, np.ceil(NUM/float(COLS)), i + 1)\n",
    "        x = np.random.randint(max_count)\n",
    "        a0, c0 = data[:, :, x], labels[x]\n",
    "        plt.imshow(a0)\n",
    "        plt.title(mapping[c0], fontsize=46)\n",
    "    fig.set_size_inches(np.array(fig.get_size_inches())*NUM)\n",
    "    #print(\"data: {}, {}\".format(data[:, :, :count].shape, labels[:count].shape))\n",
    "    ndata = np.zeros((max_count, 28, 28))\n",
    "    for i in range(max_count):\n",
    "        ndata[i, :, :] = data[:, :, i]\n",
    "    \n",
    "    fig = plt.figure()\n",
    "    x = np.random.randint(max_count)\n",
    "    #print(\"ndata: {},{},{}\".format(x, ndata[x].shape, ndata[x, :, :]))\n",
    "    a0, c0 = ndata[x, :, :], labels[x]\n",
    "    plt.imshow(a0)\n",
    "    plt.title(mapping[c0], fontsize=46)\n",
    "    \n",
    "    X_train, X_test, y_train, y_test = train_test_split(ndata[:count], labels[:count], test_size=0.1)\n",
    "    X_train, X_test = reshape_X(X_train), reshape_X(X_test)\n",
    "    return X_train, X_test, y_train, y_test\n",
    "\n",
    "X_train, X_test, y_train, y_test = generate_dataset('data/notMNIST/notMNIST_small/')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/notMNIST/notMNIST_small/\n",
      "Found 18726 files\n",
      "exception 'module' object has no attribute 'register_decoder'\n",
      "exception 'module' object has no attribute 'register_decoder'\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACYgAAAcXCAYAAABuYAfPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11262b410>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAEkCAYAAADghueSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112916550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import sys, os\n",
    "from scipy.io import savemat\n",
    "import numpy as np, glob, sys, os\n",
    "import PIL.Image as Image\n",
    "from matplotlib.pyplot import imshow\n",
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "def load_dataset():\n",
    "    #support both python 2 and 3\n",
    "    if sys.version_info[0] == 2:\n",
    "        from urllib import urlretrieve\n",
    "    else:\n",
    "        from urllib.request import urlretrieve\n",
    "\n",
    "    basedir = 'data/notMNIST/'\n",
    "    if not os.path.exists(basedir):\n",
    "        os.makedirs(basedir)\n",
    "        \n",
    "    def download(filename, source='http://yaroslavvb.com/upload/notMNIST/'):\n",
    "        localfile = os.path.join(basedir, filename)\n",
    "        if not os.path.exists(localfile):\n",
    "            print (\"Downloading %s\" % filename)\n",
    "            urlretrieve(source + filename, localfile)\n",
    "\n",
    "    def reshape_X(X_pre):\n",
    "        X_post = np.zeros((X_pre.shape[0], 1, 28, 28))\n",
    "        for i in range(X_pre.shape[0]):\n",
    "            X_post[i, 0, :, :] = X_pre[i, :, :]\n",
    "        return X_post\n",
    "\n",
    "    mapping = {}\n",
    "\n",
    "    def generate_dataset(folder):\n",
    "        print folder\n",
    "        max_count = 0\n",
    "        for (root, dirs, files) in os.walk(folder):\n",
    "            for f in files:\n",
    "                if f.endswith('.png'):\n",
    "                    max_count+=1\n",
    "\n",
    "        print 'Found %s files'%(max_count,)\n",
    "        data = np.zeros((28,28,max_count))\n",
    "        labels = np.zeros((max_count,))\n",
    "        count = 0\n",
    "        for (root, dirs, files) in os.walk(folder):\n",
    "            for f in files:\n",
    "                if f.endswith('.png'):\n",
    "                    try:\n",
    "                        #print(\"opening {}, {}, {}\".format(root, f, os.path.join(root, f)))\n",
    "                        img = Image.open(os.path.join(root, f));\n",
    "                        #print(\"{}, {}\".format(f, dirs))\n",
    "                        data[:,:,count]=np.asarray(img)\n",
    "                        surround_folder = os.path.split(root)[-1]\n",
    "                        assert len(surround_folder)==1\n",
    "                        ordinal = ord(surround_folder)-ord('A')\n",
    "                        labels[count]=ordinal\n",
    "                        mapping[ordinal] = surround_folder\n",
    "                        count+=1\n",
    "                    except Exception as e:\n",
    "                        print(\"exception {}\".format(e))\n",
    "                        pass\n",
    "\n",
    "        # Plot NUM samples\n",
    "        fig = plt.figure()\n",
    "        COLS, NUM = 3, 8\n",
    "        np.random.seed(NUM)\n",
    "        for i in range(NUM):\n",
    "            ax = fig.add_subplot(COLS, np.ceil(NUM/float(COLS)), i + 1)\n",
    "            x = np.random.randint(max_count)\n",
    "            a0, c0 = data[:, :, x], labels[x]\n",
    "            plt.imshow(a0)\n",
    "            plt.title(mapping[c0], fontsize=46)\n",
    "        fig.set_size_inches(np.array(fig.get_size_inches())*NUM)\n",
    "        #print(\"data: {}, {}\".format(data[:, :, :count].shape, labels[:count].shape))\n",
    "        ndata = np.zeros((max_count, 28, 28))\n",
    "        for i in range(max_count):\n",
    "            ndata[i, :, :] = data[:, :, i]\n",
    "\n",
    "        fig = plt.figure()\n",
    "        x = np.random.randint(max_count)\n",
    "        #print(\"ndata: {},{},{}\".format(x, ndata[x].shape, ndata[x, :, :]))\n",
    "        a0, c0 = ndata[x, :, :], labels[x]\n",
    "        plt.imshow(a0)\n",
    "        plt.title(mapping[c0], fontsize=46)\n",
    "    \n",
    "        X_train, X_test, y_train, y_test = train_test_split(ndata[:count], labels[:count], test_size=0.1)\n",
    "        X_train, X_test = reshape_X(X_train), reshape_X(X_test)\n",
    "        return X_train, X_test, y_train, y_test\n",
    "    \n",
    "    return generate_dataset('data/notMNIST/notMNIST_small/')\n",
    "\n",
    "X_train, X_test, y_train, y_test = load_dataset()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((16851, 1, 28, 28),\n",
       " (1873, 1, 28, 28),\n",
       " (16851,),\n",
       " (1873,),\n",
       " <TensorType(float64, 4D)>,\n",
       " <TensorType(float64, vector)>)"
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape, input_var, target_var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#theano shared vars\n",
    "input_var = theano.shared(X_train[:500, ...].astype(np.float64))\n",
    "target_var = theano.shared(y_train[:500, ...].astype(np.float64))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model specification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def build_ann(init):\n",
    "    l_in = lasagne.layers.InputLayer(shape=(None, 1, 28, 28), input_var=input_var)\n",
    "    l_hid1 = lasagne.layers.DenseLayer(l_in, num_units=25, nonlinearity=lasagne.nonlinearities.tanh,\n",
    "                                      b=init,\n",
    "                                      W=init)\n",
    "    l_hid2 = lasagne.layers.DenseLayer(l_hid1, num_units=25, nonlinearity=lasagne.nonlinearities.tanh,\n",
    "                                      b=init,\n",
    "                                      W=init)\n",
    "    #Final layer of 10 units, softmax across 10 labels\n",
    "    l_out = lasagne.layers.DenseLayer(l_hid2, num_units=10, nonlinearity=lasagne.nonlinearities.softmax,\n",
    "                                     b=init,\n",
    "                                     W=init)\n",
    "    prediction = lasagne.layers.get_output(l_out)\n",
    "    #categorical distribution\n",
    "    out = pm.Categorical('out', p=prediction, observed=target_var)\n",
    "    return out\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Priors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class GaussianWeights(object):\n",
    "    def __init__(self):\n",
    "        self.count = 0\n",
    "    def __call__(self, shape):\n",
    "        self.count += 1\n",
    "        return pm.Normal('w%d' % self.count, \n",
    "                         mu=0, \n",
    "                         sd=.1,\n",
    "                         testval=np.random.normal(size=shape).astype(np.float64),\n",
    "                         shape=shape)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tensors and RVs for mini-batches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#ADVI minibatches\n",
    "#\n",
    "from six.moves import zip\n",
    "minibatch_tensors = [input_var, target_var]\n",
    "def create_minibatch(data, batchsize=500):\n",
    "    rng = np.random.RandomState(0)\n",
    "    start_idx = 0\n",
    "    #Return random samples of batchsize on each iteration\n",
    "    while True:\n",
    "        ixs = rng.randint(data.shape[0], size=batchsize)\n",
    "        yield data[ixs]\n",
    "\n",
    "minibatches = zip(\n",
    "    create_minibatch(X_train, 500), \n",
    "    create_minibatch(y_train, 500)\n",
    ")\n",
    "total_size = len(y_train)\n",
    "\n",
    "def run_advi(likelihood, advi_iters=50000):\n",
    "    #Train\n",
    "    input_var.set_value(X_train[:500, ...])\n",
    "    target_var.set_value(y_train[:500, ...])\n",
    "    v_params = pm.variational.advi_minibatch(\n",
    "        n=advi_iters,\n",
    "        minibatch_tensors=minibatch_tensors,\n",
    "        minibatch_RVs=[likelihood], \n",
    "        minibatches=minibatches,\n",
    "        total_size=total_size,\n",
    "        learning_rate=1e-2,\n",
    "        epsilon=1.0\n",
    "    )\n",
    "\n",
    "    trace = pm.variational.sample_vp(v_params, draws=500)\n",
    "    \n",
    "    #Predict on test data\n",
    "    input_var.set_value(X_test)\n",
    "    target_var.set_value(y_test)\n",
    "    \n",
    "    ppc = pm.sample_ppc(trace, samples=100)\n",
    "    y_pred = mode(ppc['out'], axis=0).mode[0, :]\n",
    "    \n",
    "    return v_params, trace, ppc, y_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now put them together"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Average ELBO = -20,222.51: 100%|██████████| 50000/50000 [05:38<00:00, 156.82it/s] \n",
      "Finished minibatch ADVI: ELBO = -21,520.57\n",
      "/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/ipykernel/__main__.py:33: DeprecationWarning: Old ADVI interface and sample_vp is deprecated and will be removed in future, use pm.fit and pm.sample_approx instead\n",
      "100%|██████████| 500/500 [00:00<00:00, 986.94it/s] \n",
      "100%|██████████| 100/100 [00:03<00:00, 25.56it/s]\n"
     ]
    }
   ],
   "source": [
    "with pm.Model() as neural_net:\n",
    "    likelihood = build_ann(GaussianWeights())\n",
    "    v_params, trace, ppc, y_pred = run_advi(likelihood)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaMAAAD8CAYAAADaOstiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFEYiIhKcwkhERIJTGImISHD/H6j59wCOT+fUAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1235f1c90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(v_params.elbo_vals)\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x123583b90>"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWkAAAEBCAYAAAC63FR5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWYX9F7a81KSUwcRHAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1235f1b50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on test: 81.6337426588\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy on test: {}\".format(accuracy_score(y_test, y_pred)*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hierarchical Neural Network: Learning Regularization from data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((16851, 1, 28, 28), (1873, 1, 28, 28), (16851,), (1873,))"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from six.moves import zip\n",
    "class GaussWeightWithHR(object):\n",
    "    def __init__(self):\n",
    "        self.count = 0\n",
    "    def __call__(self, shape):\n",
    "        self.count += 1\n",
    "        regularization = pm.HalfNormal('reg_hyper%d' % self.count, sd=1)\n",
    "        return pm.Normal('w%d'%self.count, \n",
    "                         mu=0, \n",
    "                         sd=regularization, \n",
    "                         testval=np.random.normal(size=shape),\n",
    "                         shape=shape)\n",
    "minibatches = zip(\n",
    "    create_minibatch(X_train, 500),\n",
    "    create_minibatch(y_train, 500)\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def run_fit_sample(likelihood, advi_iters=50000):\n",
    "    #Train\n",
    "    input_var.set_value(X_train[:500, ...])\n",
    "    target_var.set_value(y_train[:500, ...])\n",
    "    #v_params = pm.fit(n=20)\n",
    "\n",
    "    trace = pm.sample()\n",
    "    \n",
    "    #Predict on test data\n",
    "    input_var.set_value(X_test)\n",
    "    target_var.set_value(y_test)\n",
    "    \n",
    "    ppc = pm.sample_ppc(trace, samples=100)\n",
    "    y_pred = mode(ppc['out'], axis=0).mode[0, :]\n",
    "    \n",
    "    return v_params, trace, ppc, y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "100%|██████████| 1000/1000 [1:01:37<00:00,  3.85s/it]/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:459: UserWarning: Chain 0 reached the maximum tree depth. Increase max_treedepth, increase target_accept or reparameterize.\n",
      "  'reparameterize.' % self._chain_id)\n",
      "/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:451: UserWarning: The acceptance probability in chain 0 does not match the target. It is 0.665976970964, but should be close to 0.8. Try to increase the number of tuning steps.\n",
      "  % (self._chain_id, mean_accept, target_accept))\n",
      "\n",
      "100%|██████████| 100/100 [00:04<00:00, 25.31it/s]\n"
     ]
    }
   ],
   "source": [
    "with pm.Model() as hnn:\n",
    "    likelihood = build_ann(GaussWeightWithHR())\n",
    "    v_params, trace, ppc, y_pred = run_fit_sample(likelihood)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaMAAAD8CAYAAADaOstiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a046150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(v_params.elbo_vals)\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on test: 62.9471436199\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy on test: {}\".format(accuracy_score(y_test, y_pred)*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x12e234b10>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x122f34290>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x12e14cbd0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1174b9610>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x122f35f10>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x126f09590>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x126a55f10>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x120b365d0>]], dtype=object)"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAAI4CAYAAAB3HEhGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKIqiKIqiKMoAUR0sRVEURVEURVGUAfJfYLhKmFu+gP4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x126e4fa90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pm.traceplot(trace, varnames=['reg_hyper1', 'reg_hyper2', 'reg_hyper3', 'w2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['reg_hyper1_log__',\n",
       " 'w1',\n",
       " 'reg_hyper2_log__',\n",
       " 'w2',\n",
       " 'reg_hyper3_log__',\n",
       " 'w3',\n",
       " 'reg_hyper4_log__',\n",
       " 'w4',\n",
       " 'reg_hyper5_log__',\n",
       " 'w5',\n",
       " 'reg_hyper6_log__',\n",
       " 'w6',\n",
       " 'reg_hyper1',\n",
       " 'reg_hyper2',\n",
       " 'reg_hyper3',\n",
       " 'reg_hyper4',\n",
       " 'reg_hyper5',\n",
       " 'reg_hyper6']"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trace.varnames"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now a Convolutional Net\n",
    "All it requires is that we change the build_ann method to include convolutional layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def build_ann_conv(init):\n",
    "    network = lasagne.layers.InputLayer(shape=(None, 1, 28, 28), input_var=input_var)\n",
    "    network = lasagne.layers.Conv2DLayer(network, \n",
    "                                      num_filters=32, \n",
    "                                      filter_size=(5,5),\n",
    "                                      nonlinearity=lasagne.nonlinearities.tanh,\n",
    "                                      W=init\n",
    "                                     )\n",
    "    network = lasagne.layers.MaxPool2DLayer(network, pool_size=(2,2))\n",
    "    network = lasagne.layers.Conv2DLayer(network,\n",
    "                                        num_filters=32,\n",
    "                                        filter_size=(5,5),\n",
    "                                        nonlinearity=lasagne.nonlinearities.tanh,\n",
    "                                        W=init)\n",
    "    network = lasagne.layers.MaxPool2DLayer(network, pool_size=(2,2))\n",
    "    network = lasagne.layers.DenseLayer(network, \n",
    "                                        num_units=256, \n",
    "                                        nonlinearity=lasagne.nonlinearities.tanh,\n",
    "                                        b=init,\n",
    "                                        W=init\n",
    "                                       )\n",
    "    network = lasagne.layers.DenseLayer(network,\n",
    "                                       num_units=10,\n",
    "                                       nonlinearity=lasagne.nonlinearities.softmax,\n",
    "                                       b=init,\n",
    "                                       W=init)\n",
    "    prediction = lasagne.layers.get_output(network)\n",
    "    \n",
    "    return pm.Categorical('out', prediction, observed=target_var)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "  6%|▌         | 55/1000 [2:51:55<120:21:50, 458.53s/it]/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:429: UserWarning: Chain 0 contains only 0 samples.\n",
      "  % (self._chain_id, n))\n",
      "/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:431: UserWarning: Step size tuning was enabled throughout the whole trace. You might want to specify the number of tuning steps.\n",
      "  warnings.warn('Step size tuning was enabled throughout the whole '\n",
      "/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/numpy/core/fromnumeric.py:2909: RuntimeWarning: Mean of empty slice.\n",
      "  out=out, **kwargs)\n",
      "\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "low >= high",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m\u001b[0m",
      "\u001b[0;31mValueError\u001b[0mTraceback (most recent call last)",
      "\u001b[0;32m<ipython-input-178-071c87a84bbc>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mpm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mModel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mcnn\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m     \u001b[0mlikelihood\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbuild_ann_conv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mGaussWeightWithHR\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m     \u001b[0mv_params\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mppc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrun_fit_sample\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlikelihood\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-165-daf9ec12e3cc>\u001b[0m in \u001b[0;36mrun_fit_sample\u001b[0;34m(likelihood, advi_iters)\u001b[0m\n\u001b[1;32m     11\u001b[0m     \u001b[0mtarget_var\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_value\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 13\u001b[0;31m     \u001b[0mppc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msample_ppc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msamples\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m100\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     14\u001b[0m     \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mppc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'out'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/vmullachery/anaconda/envs/dl2.7/lib/python2.7/site-packages/pymc3/sampling.pyc\u001b[0m in \u001b[0;36msample_ppc\u001b[0;34m(trace, samples, model, vars, size, random_seed, progressbar)\u001b[0m\n\u001b[1;32m    561\u001b[0m     \u001b[0mseed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrandom_seed\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    562\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 563\u001b[0;31m     \u001b[0mindices\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrandint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msamples\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    564\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mprogressbar\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    565\u001b[0m         \u001b[0mindices\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtqdm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtotal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msamples\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mmtrand.pyx\u001b[0m in \u001b[0;36mmtrand.RandomState.randint\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: low >= high"
     ]
    }
   ],
   "source": [
    "with pm.Model() as cnn:\n",
    "    likelihood = build_ann_conv(GaussWeightWithHR())\n",
    "    v_params, trace, ppc, y_pred = run_fit_sample(likelihood)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on test: 62.9471436199\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy on test: {}\".format(accuracy_score(y_test, y_pred)*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Unpredictability on categorical distribution measured using Chi-squared statistic\n",
    "We don't simply use the ppc standard deviation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "miss_class = np.where(y_test != y_pred)[0]\n",
    "corr_class = np.where(y_test == y_pred)[0]\n",
    "preds = pd.DataFrame(ppc['out']).T\n",
    "chis = preds.apply(lambda x: chisquare(x).statistic, axis='columns')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11f767250>"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEOCAYAAACNY7BQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11f6e0a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(chis.loc[miss_class].dropna(), label='Error')\n",
    "sns.distplot(chis.loc[corr_class].dropna(), label='Correct')\n",
    "plt.legend()\n",
    "sns.despine()\n",
    "plt.xlabel('Chi-Square Statistic')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}