diff --git a/notebooks/week-1/week 1 - VM test.ipynb b/notebooks/week-1/week 1 - VM test.ipynb
index 235273c..508c16c 100644
--- a/notebooks/week-1/week 1 - VM test.ipynb	
+++ b/notebooks/week-1/week 1 - VM test.ipynb	
@@ -11,22 +11,63 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "hello world\n"
+     ]
+    }
+   ],
    "source": [
     "print \"hello world\""
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "keras successfully imported\n",
+      "theano successfully imported\n",
+      "tensorflow successfully imported\n",
+      "sklearn successfully imported\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/vagrant/anaconda2/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.\n",
+      "  warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "seaborn successfully imported\n"
+     ]
+    }
+   ],
    "source": [
     "import keras\n",
     "print \"keras successfully imported\"\n",
@@ -42,13 +83,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "uni: jl4745\n"
+     ]
+    }
+   ],
    "source": [
-    "print \"uni:\", \"fill in your uni here\""
+    "print \"uni:\", \"jl4745\""
    ]
   }
  ],
diff --git a/notebooks/week-2/lab2.ipynb b/notebooks/week-2/lab2.ipynb
new file mode 100644
index 0000000..2728074
--- /dev/null
+++ b/notebooks/week-2/lab2.ipynb
@@ -0,0 +1,186 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "start game 0\n",
+      "player0Win,7vs2\n",
+      "player1Lose,0vs7\n",
+      "player2Lose,3vs5\n",
+      "player3Lose,3vs9\n",
+      "player4Lose,0vs4\n",
+      "\n",
+      "start game 1\n",
+      "player0Win,2vs0\n",
+      "player1Win,6vs1\n",
+      "player2Lose,3vs4\n",
+      "player3Win,5vs4\n",
+      "player4Win,2vs1\n",
+      "\n",
+      "start game 2\n",
+      "player0Win,6vs1\n",
+      "player1Win,6vs1\n",
+      "player2Lose,3vs7\n",
+      "player3Win,9vs5\n",
+      "player4Win,9vs1\n",
+      "\n",
+      "start game 3\n",
+      "player0Win,9vs8\n",
+      "player1Win,6vs3\n",
+      "player2Lose,1vs5\n",
+      "player3Win,4vs2\n",
+      "player4Lose,4vs6\n",
+      "\n",
+      "start game 4\n",
+      "player0Win,8vs0\n",
+      "player1Lose,3vs8\n",
+      "player2Win,6vs5\n",
+      "player3Lose,5vs6\n",
+      "player4Win,3vs0\n",
+      "\n",
+      "start game 5\n",
+      "player0Win,4vs3\n",
+      "player1Lose,2vs8\n",
+      "player2Lose,0vs2\n",
+      "player3Lose,3vs9\n",
+      "player4Lose,1vs2\n",
+      "\n",
+      "start game 6\n",
+      "player0Win,9vs0\n",
+      "player1Win,2vs2\n",
+      "player2Win,5vs3\n",
+      "player3Lose,0vs1\n",
+      "player4Lose,2vs7\n",
+      "\n",
+      "start game 7\n",
+      "player0Lose,2vs7\n",
+      "player1Win,8vs2\n",
+      "player2Win,5vs2\n",
+      "player3Lose,3vs5\n",
+      "player4Win,0vs0\n",
+      "\n",
+      "start game 8\n",
+      "player0Win,9vs4\n",
+      "player1Win,3vs3\n",
+      "player2Win,6vs2\n",
+      "player3Lose,8vs9\n",
+      "player4Win,9vs1\n",
+      "\n",
+      "start game 9\n",
+      "player0Lose,0vs4\n",
+      "player1Lose,3vs6\n",
+      "player2Lose,1vs6\n",
+      "player3Win,8vs2\n",
+      "player4Win,6vs6\n",
+      "\n",
+      "game result:\n",
+      "player0has $800left.\n",
+      "player1has $600left.\n",
+      "player2has $400left.\n",
+      "player3has $400left.\n",
+      "player4has $600left.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import random\n",
+    "\n",
+    "gameStake = 50\n",
+    "cards = range(10)\n",
+    "\n",
+    "class Player:\n",
+    "    \n",
+    "    playerID=0\n",
+    "    playerPot=0\n",
+    "    \n",
+    "    def __init__(self, inputID, startingPot):\n",
+    "        self.playerID = inputID\n",
+    "        self.playerPot = startingPot\n",
+    "        \n",
+    "    def play(self, dealerCard):\n",
+    "        \n",
+    "        playerCard = random.choice(cards)\n",
+    "        \n",
+    "        if playerCard < dealerCard:\n",
+    "            self.playerPot += -gameStake\n",
+    "            print 'player'+ str(self.playerID) + 'Lose,'+ str(playerCard)+'vs'+ str(dealerCard)\n",
+    "        else:\n",
+    "            self.playerPot += gameStake\n",
+    "            print 'player'+ str(self.playerID) + 'Win,'+ str(playerCard)+'vs'+ str(dealerCard)\n",
+    "            \n",
+    "    def returnPot(self):\n",
+    "        return self.playerPot\n",
+    "    \n",
+    "    def returnID(self):\n",
+    "        return self.playerID\n",
+    "    \n",
+    "    \n",
+    "def playHand(players):\n",
+    "        \n",
+    "        for player in players:\n",
+    "            dealerCard = random.choice(cards)\n",
+    "            player.play(dealerCard)\n",
+    "            \n",
+    "def checkBalances(players):\n",
+    "        \n",
+    "        for player in players:\n",
+    "            \n",
+    "            print 'player' + str(player.returnID()) +'has $'+ str(player.returnPot())+'left.'\n",
+    "            \n",
+    "players =[]\n",
+    "\n",
+    "for i in range(5):\n",
+    "    players.append(Player(i,500))\n",
+    "    \n",
+    "for i in range(10):\n",
+    "    print ''\n",
+    "    print 'start game ' + str(i)\n",
+    "    playHand(players)\n",
+    "    \n",
+    "print ''\n",
+    "print 'game result:'\n",
+    "checkBalances(players)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-3/lab3.ipynb b/notebooks/week-3/lab3.ipynb
new file mode 100644
index 0000000..9e1d208
--- /dev/null
+++ b/notebooks/week-3/lab3.ipynb
@@ -0,0 +1,616 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "%matplotlib inline\n",
+    "import random\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns; sns.set(style=\"ticks\", color_codes=True)\n",
+    "from sklearn.preprocessing import OneHotEncoder\n",
+    "from sklearn.utils import shuffle\n",
+    "\n",
+    "class Network(object):\n",
+    "    \n",
+    "    def __init__(self, sizes):\n",
+    "        \n",
+    "        \"\"\"The list ``sizes`` contains the number of neurons in the\n",
+    "        respective layers of the network.  For example, if the list\n",
+    "        was [2, 3, 1] then it would be a three-layer network, with the\n",
+    "        first layer containing 2 neurons, the second layer 3 neurons,\n",
+    "        and the third layer 1 neuron.  The biases and weights for the\n",
+    "        network are initialized randomly, using a Gaussian\n",
+    "        distribution with mean 0, and variance 1.  Note that the first\n",
+    "        layer is assumed to be an input layer, and by convention we\n",
+    "        won't set any biases for those neurons, since biases are only\n",
+    "        ever used in computing the outputs for later layers.\"\"\"\n",
+    "        \n",
+    "        self.num_layers = len(sizes)\n",
+    "        self.sizes = sizes\n",
+    "        self.biases = [np.random.randn(y, 1) for y in sizes[1:]]\n",
+    "        self.weights = [np.random.randn(y, x) for x, y in zip(sizes[:-1], sizes[1:])]\n",
+    "        \n",
+    "    def feedforward (self, a):\n",
+    "        \n",
+    "        \"\"\"Return the output of the network if \"a\" is input. The np.dot() \n",
+    "        function computes the matrix multiplication between the weight and input\n",
+    "        matrices for each set of layers. When used with numpy arrays, the '+'\n",
+    "        operator performs matrix addition.\"\"\"\n",
+    "        \n",
+    "        for b, w in zip(self.biases, self.weights):\n",
+    "            a = sigmoid(np.dot(w, a)+b)\n",
+    "        return a\n",
+    "    \n",
+    "    def SGD(self, training_data, epochs, mini_batch_size, eta, test_data=None):\n",
+    "        \n",
+    "        \"\"\"Train the neural network using mini-batch stochastic\n",
+    "        gradient descent.  The \"training_data\" is a list of tuples\n",
+    "        \"(x, y)\" representing the training inputs and the desired\n",
+    "        outputs.  The other non-optional parameters specify the number \n",
+    "        of epochs, size of each mini-batch, and the learning rate.  \n",
+    "        If \"test_data\" is provided then the network will be evaluated \n",
+    "        against the test data after each epoch, and partial progress \n",
+    "        printed out.  This is useful for tracking progress, but slows\n",
+    "        things down substantially.\"\"\"\n",
+    "        \n",
+    "        # create an empty array to store the accuracy results from each epoch\n",
+    "        results = []\n",
+    "\n",
+    "        n = len(training_data)\n",
+    "        \n",
+    "        if test_data: \n",
+    "            n_test = len(test_data)\n",
+    "            \n",
+    "        # this is the code for one training step, done once for each epoch\n",
+    "        for j in xrange(epochs):\n",
+    "            \n",
+    "            # before each epoch, the data is randomly shuffled\n",
+    "            random.shuffle(training_data)\n",
+    "            \n",
+    "            # training data is broken up into individual mini-batches\n",
+    "            mini_batches = [ training_data[k:k+mini_batch_size] \n",
+    "                            for k in xrange(0, n, mini_batch_size) ]\n",
+    "            \n",
+    "            # then each mini-batch is used to update the parameters of the \n",
+    "            # network using backpropagation and the specified learning rate\n",
+    "            for mini_batch in mini_batches:\n",
+    "                self.update_mini_batch(mini_batch, eta)\n",
+    "            \n",
+    "            # if a test data set is provided, the accuracy results \n",
+    "            # are displayed and stored in the 'results' array\n",
+    "            if test_data:\n",
+    "                num_correct = self.evaluate(test_data)\n",
+    "                accuracy = \"%.2f\" % (100 * (float(num_correct) / n_test))\n",
+    "                print \"Epoch\", j, \":\", num_correct, \"/\", n_test, \"-\", accuracy, \"% acc\"\n",
+    "                results.append(accuracy)\n",
+    "            else:\n",
+    "                print \"Epoch\", j, \"complete\"\n",
+    "                \n",
+    "        return results\n",
+    "    \n",
+    "    def update_mini_batch(self, mini_batch, eta):\n",
+    "        \n",
+    "        \"\"\"Update the network's weights and biases by applying\n",
+    "        gradient descent using backpropagation to a single mini batch.\n",
+    "        The \"mini_batch\" is a list of tuples \"(x, y)\", and \"eta\"\n",
+    "        is the learning rate.\"\"\"\n",
+    "\n",
+    "        nabla_b = [np.zeros(b.shape) for b in self.biases]\n",
+    "        nabla_w = [np.zeros(w.shape) for w in self.weights]\n",
+    "        for x, y in mini_batch:\n",
+    "            delta_nabla_b, delta_nabla_w = self.backprop(x, y)\n",
+    "            nabla_b = [nb+dnb for nb, dnb in zip(nabla_b, delta_nabla_b)]\n",
+    "            nabla_w = [nw+dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]\n",
+    "        self.weights = [w-(eta/len(mini_batch))*nw \n",
+    "                        for w, nw in zip(self.weights, nabla_w)]\n",
+    "        self.biases = [b-(eta/len(mini_batch))*nb \n",
+    "                       for b, nb in zip(self.biases, nabla_b)]\n",
+    "        \n",
+    "    def backprop(self, x, y):\n",
+    "        \n",
+    "        \"\"\"Return a tuple ``(nabla_b, nabla_w)`` representing the\n",
+    "        gradient for the cost function C_x.  ``nabla_b`` and\n",
+    "        ``nabla_w`` are layer-by-layer lists of numpy arrays, similar\n",
+    "        to ``self.biases`` and ``self.weights``.\"\"\"\n",
+    "        \n",
+    "        nabla_b = [np.zeros(b.shape) for b in self.biases]\n",
+    "        nabla_w = [np.zeros(w.shape) for w in self.weights]\n",
+    "        \n",
+    "        # feedforward\n",
+    "        activation = x\n",
+    "        activations = [x] # list to store all the activations, layer by layer\n",
+    "        zs = [] # list to store all the z vectors, layer by layer\n",
+    "        for b, w in zip(self.biases, self.weights):\n",
+    "            z = np.dot(w, activation)+b\n",
+    "            zs.append(z)\n",
+    "            activation = sigmoid(z)\n",
+    "            activations.append(activation)\n",
+    "        # backward pass\n",
+    "        delta = self.cost_derivative(activations[-1], y) * \\\n",
+    "            sigmoid_prime(zs[-1])\n",
+    "        nabla_b[-1] = delta\n",
+    "        nabla_w[-1] = np.dot(delta, activations[-2].transpose())\n",
+    "        \n",
+    "        \"\"\"Note that the variable l in the loop below is used a little\n",
+    "        differently to the notation in Chapter 2 of the book.  Here,\n",
+    "        l = 1 means the last layer of neurons, l = 2 is the\n",
+    "        second-last layer, and so on.  It's a renumbering of the\n",
+    "        scheme in the book, used here to take advantage of the fact\n",
+    "        that Python can use negative indices in lists.\"\"\"\n",
+    "        \n",
+    "        for l in xrange(2, self.num_layers):\n",
+    "            z = zs[-l]\n",
+    "            sp = sigmoid_prime(z)\n",
+    "            delta = np.dot(self.weights[-l+1].transpose(), delta) * sp\n",
+    "            nabla_b[-l] = delta\n",
+    "            nabla_w[-l] = np.dot(delta, activations[-l-1].transpose())\n",
+    "        return (nabla_b, nabla_w)\n",
+    "\n",
+    "    def evaluate(self, test_data):\n",
+    "        \n",
+    "        \"\"\"Return the number of test inputs for which the neural\n",
+    "        network outputs the correct result. Note that the neural\n",
+    "        network's output is assumed to be the index of whichever\n",
+    "        neuron in the final layer has the highest activation.\n",
+    "        \n",
+    "        Numpy's argmax() function returns the position of the \n",
+    "        largest element in an array. We first create a list of \n",
+    "        predicted value and target value pairs, and then count \n",
+    "        the number of times those values match to get the total \n",
+    "        number correct.\"\"\"\n",
+    "        \n",
+    "        test_results = [(np.argmax(self.feedforward(x)), y)\n",
+    "                        for (x, y) in test_data]\n",
+    "        \n",
+    "        return sum(int(x == y) for (x, y) in test_results)\n",
+    "\n",
+    "    def cost_derivative(self, output_activations, y):\n",
+    "        \"\"\"Return the vector of partial derivatives \\partial C_x /\n",
+    "        \\partial a for the output activations.\"\"\"\n",
+    "        return (output_activations-y)\n",
+    "    \n",
+    "def sigmoid(z):\n",
+    "#     The sigmoid activation function.\n",
+    "    return 1.0/(1.0 + np.exp(-z))\n",
+    "\n",
+    "def sigmoid_prime(z):\n",
+    "#     Derivative of the sigmoid function.\n",
+    "    return sigmoid(z)*(1-sigmoid(z))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "     sepal_length  sepal_width  petal_length  petal_width     species\n",
+      "49            5.0          3.3           1.4          0.2      setosa\n",
+      "44            5.1          3.8           1.9          0.4      setosa\n",
+      "83            6.0          2.7           5.1          1.6  versicolor\n",
+      "5             5.4          3.9           1.7          0.4      setosa\n",
+      "124           6.7          3.3           5.7          2.1   virginica\n",
+      "[[ 0.63291139  0.75        0.20289855  0.08      ]\n",
+      " [ 0.64556962  0.86363636  0.27536232  0.16      ]\n",
+      " [ 0.75949367  0.61363636  0.73913043  0.64      ]\n",
+      " [ 0.6835443   0.88636364  0.24637681  0.16      ]\n",
+      " [ 0.84810127  0.75        0.82608696  0.84      ]\n",
+      " [ 0.62025316  0.70454545  0.2173913   0.04      ]\n",
+      " [ 0.91139241  0.81818182  0.88405797  1.        ]\n",
+      " [ 0.87341772  0.72727273  0.82608696  0.92      ]\n",
+      " [ 0.88607595  0.72727273  0.68115942  0.56      ]\n",
+      " [ 0.64556962  0.79545455  0.20289855  0.12      ]\n",
+      " [ 0.84810127  0.70454545  0.63768116  0.56      ]\n",
+      " [ 0.58227848  0.77272727  0.20289855  0.12      ]\n",
+      " [ 0.75949367  0.68181818  0.69565217  0.72      ]\n",
+      " [ 0.79746835  0.56818182  0.71014493  0.6       ]\n",
+      " [ 0.82278481  0.68181818  0.79710145  0.72      ]\n",
+      " [ 1.          0.86363636  0.92753623  0.8       ]\n",
+      " [ 0.63291139  0.72727273  0.17391304  0.08      ]\n",
+      " [ 0.69620253  0.56818182  0.57971014  0.52      ]\n",
+      " [ 0.63291139  0.45454545  0.50724638  0.4       ]\n",
+      " [ 0.63291139  0.77272727  0.2173913   0.08      ]\n",
+      " [ 0.60759494  0.70454545  0.23188406  0.08      ]\n",
+      " [ 0.78481013  0.5         0.65217391  0.6       ]\n",
+      " [ 0.63291139  0.77272727  0.23188406  0.16      ]\n",
+      " [ 0.7721519   0.68181818  0.71014493  0.72      ]\n",
+      " [ 0.79746835  0.77272727  0.8115942   0.96      ]\n",
+      " [ 0.70886076  0.68181818  0.5942029   0.52      ]\n",
+      " [ 0.82278481  0.68181818  0.75362319  0.8       ]\n",
+      " [ 0.84810127  0.70454545  0.68115942  0.6       ]\n",
+      " [ 0.58227848  0.72727273  0.20289855  0.08      ]\n",
+      " [ 0.60759494  0.77272727  0.27536232  0.08      ]\n",
+      " [ 0.65822785  0.77272727  0.20289855  0.08      ]\n",
+      " [ 0.6835443   0.77272727  0.2173913   0.16      ]\n",
+      " [ 0.64556962  0.86363636  0.23188406  0.08      ]\n",
+      " [ 0.79746835  0.61363636  0.71014493  0.72      ]\n",
+      " [ 0.62025316  0.56818182  0.65217391  0.68      ]\n",
+      " [ 0.6835443   0.88636364  0.1884058   0.16      ]\n",
+      " [ 0.72151899  0.86363636  0.24637681  0.12      ]\n",
+      " [ 0.75949367  0.77272727  0.65217391  0.64      ]\n",
+      " [ 0.64556962  0.77272727  0.2173913   0.08      ]\n",
+      " [ 0.64556962  0.84090909  0.2173913   0.16      ]\n",
+      " [ 0.70886076  0.63636364  0.71014493  0.8       ]\n",
+      " [ 0.74683544  0.72727273  0.69565217  0.72      ]\n",
+      " [ 0.84810127  0.68181818  0.75362319  0.92      ]\n",
+      " [ 0.86075949  0.68181818  0.79710145  0.84      ]\n",
+      " [ 0.82278481  0.63636364  0.66666667  0.6       ]\n",
+      " [ 0.7721519   0.63636364  0.68115942  0.48      ]\n",
+      " [ 0.7721519   0.63636364  0.57971014  0.52      ]\n",
+      " [ 0.58227848  0.70454545  0.2173913   0.08      ]\n",
+      " [ 0.64556962  0.86363636  0.2173913   0.12      ]\n",
+      " [ 0.87341772  0.70454545  0.73913043  0.92      ]\n",
+      " [ 0.92405063  0.65909091  0.91304348  0.72      ]\n",
+      " [ 0.7721519   0.65909091  0.68115942  0.56      ]\n",
+      " [ 0.73417722  0.61363636  0.5942029   0.4       ]\n",
+      " [ 0.78481013  0.65909091  0.62318841  0.52      ]\n",
+      " [ 0.63291139  0.81818182  0.20289855  0.08      ]\n",
+      " [ 0.79746835  0.52272727  0.63768116  0.52      ]\n",
+      " [ 0.78481013  0.63636364  0.69565217  0.72      ]\n",
+      " [ 0.81012658  0.63636364  0.8115942   0.84      ]\n",
+      " [ 0.97468354  0.63636364  0.97101449  0.8       ]\n",
+      " [ 0.75949367  0.65909091  0.65217391  0.6       ]\n",
+      " [ 0.86075949  0.72727273  0.85507246  0.92      ]\n",
+      " [ 0.83544304  0.68181818  0.63768116  0.56      ]\n",
+      " [ 0.81012658  0.72727273  0.65217391  0.6       ]\n",
+      " [ 0.70886076  0.65909091  0.52173913  0.52      ]\n",
+      " [ 0.82278481  0.72727273  0.73913043  0.8       ]\n",
+      " [ 0.62025316  0.81818182  0.20289855  0.04      ]\n",
+      " [ 0.63291139  0.52272727  0.47826087  0.4       ]\n",
+      " [ 0.60759494  0.68181818  0.20289855  0.12      ]\n",
+      " [ 0.91139241  0.68181818  0.84057971  0.64      ]\n",
+      " [ 0.81012658  0.70454545  0.79710145  0.72      ]\n",
+      " [ 0.63291139  0.68181818  0.23188406  0.08      ]\n",
+      " [ 0.84810127  0.70454545  0.8115942   0.96      ]\n",
+      " [ 0.97468354  0.68181818  0.88405797  0.92      ]\n",
+      " [ 0.63291139  0.79545455  0.23188406  0.24      ]\n",
+      " [ 0.89873418  0.68181818  0.85507246  0.84      ]\n",
+      " [ 0.58227848  0.81818182  0.14492754  0.08      ]\n",
+      " [ 0.82278481  0.68181818  0.84057971  0.88      ]\n",
+      " [ 0.64556962  0.75        0.24637681  0.2       ]\n",
+      " [ 0.7721519   0.59090909  0.8115942   0.56      ]\n",
+      " [ 0.72151899  0.65909091  0.60869565  0.52      ]\n",
+      " [ 0.75949367  0.5         0.72463768  0.6       ]\n",
+      " [ 0.84810127  0.68181818  0.72463768  0.68      ]\n",
+      " [ 0.93670886  0.63636364  0.88405797  0.76      ]\n",
+      " [ 0.59493671  0.72727273  0.1884058   0.08      ]\n",
+      " [ 0.72151899  0.68181818  0.60869565  0.48      ]\n",
+      " [ 0.67088608  0.84090909  0.2173913   0.08      ]\n",
+      " [ 0.73417722  0.61363636  0.73913043  0.76      ]\n",
+      " [ 0.64556962  0.79545455  0.20289855  0.08      ]\n",
+      " [ 0.79746835  0.63636364  0.73913043  0.6       ]\n",
+      " [ 0.72151899  1.          0.2173913   0.16      ]\n",
+      " [ 0.60759494  0.77272727  0.23188406  0.08      ]\n",
+      " [ 0.73417722  0.61363636  0.56521739  0.48      ]\n",
+      " [ 0.83544304  0.65909091  0.66666667  0.52      ]\n",
+      " [ 0.69620253  0.95454545  0.20289855  0.08      ]\n",
+      " [ 0.65822785  0.61363636  0.56521739  0.56      ]\n",
+      " [ 0.87341772  0.70454545  0.71014493  0.6       ]\n",
+      " [ 0.87341772  0.70454545  0.7826087   0.84      ]\n",
+      " [ 0.97468354  0.86363636  0.97101449  0.88      ]\n",
+      " [ 0.65822785  0.93181818  0.2173913   0.04      ]\n",
+      " [ 0.55696203  0.72727273  0.1884058   0.08      ]\n",
+      " [ 0.73417722  0.61363636  0.73913043  0.76      ]\n",
+      " [ 0.73417722  0.63636364  0.73913043  0.96      ]\n",
+      " [ 0.74683544  0.68181818  0.60869565  0.6       ]\n",
+      " [ 0.70886076  0.56818182  0.56521739  0.44      ]\n",
+      " [ 0.79746835  0.75        0.86956522  1.        ]\n",
+      " [ 0.5443038   0.68181818  0.15942029  0.04      ]\n",
+      " [ 0.79746835  0.65909091  0.8115942   0.72      ]\n",
+      " [ 0.69620253  0.52272727  0.57971014  0.52      ]\n",
+      " [ 0.81012658  0.61363636  0.76811594  0.76      ]\n",
+      " [ 0.55696203  0.68181818  0.1884058   0.08      ]\n",
+      " [ 0.72151899  0.63636364  0.5942029   0.52      ]\n",
+      " [ 0.75949367  0.5         0.57971014  0.4       ]\n",
+      " [ 0.70886076  0.61363636  0.60869565  0.52      ]\n",
+      " [ 0.81012658  0.63636364  0.8115942   0.88      ]\n",
+      " [ 0.6835443   0.68181818  0.65217391  0.6       ]\n",
+      " [ 0.72151899  0.59090909  0.50724638  0.4       ]\n",
+      " [ 0.84810127  0.75        0.82608696  1.        ]\n",
+      " [ 0.73417722  0.59090909  0.57971014  0.48      ]\n",
+      " [ 0.65822785  0.79545455  0.2173913   0.08      ]\n",
+      " [ 0.72151899  0.56818182  0.72463768  0.8       ]\n",
+      " [ 0.72151899  0.63636364  0.65217391  0.52      ]\n",
+      " [ 0.97468354  0.59090909  1.          0.92      ]\n",
+      " [ 0.69620253  0.54545455  0.53623188  0.4       ]\n",
+      " [ 0.62025316  0.54545455  0.47826087  0.4       ]\n",
+      " [ 0.79746835  0.75        0.68115942  0.64      ]\n",
+      " [ 0.96202532  0.68181818  0.95652174  0.84      ]\n",
+      " [ 0.6835443   0.77272727  0.24637681  0.08      ]\n",
+      " [ 0.62025316  0.68181818  0.20289855  0.08      ]\n",
+      " [ 0.84810127  0.56818182  0.84057971  0.72      ]\n",
+      " [ 0.74683544  0.68181818  0.73913043  0.72      ]\n",
+      " [ 0.69620253  0.54545455  0.55072464  0.44      ]\n",
+      " [ 0.70886076  0.68181818  0.65217391  0.6       ]\n",
+      " [ 0.81012658  0.65909091  0.62318841  0.52      ]\n",
+      " [ 0.78481013  0.77272727  0.7826087   0.92      ]\n",
+      " [ 0.59493671  0.72727273  0.23188406  0.08      ]\n",
+      " [ 0.79746835  0.56818182  0.72463768  0.76      ]\n",
+      " [ 0.81012658  0.72727273  0.76811594  0.92      ]\n",
+      " [ 0.60759494  0.68181818  0.20289855  0.04      ]\n",
+      " [ 0.91139241  0.72727273  0.86956522  0.72      ]\n",
+      " [ 0.55696203  0.65909091  0.20289855  0.08      ]\n",
+      " [ 0.6835443   0.84090909  0.2173913   0.08      ]\n",
+      " [ 0.7721519   0.68181818  0.66666667  0.56      ]\n",
+      " [ 0.69620253  0.59090909  0.63768116  0.48      ]\n",
+      " [ 0.56962025  0.52272727  0.1884058   0.12      ]\n",
+      " [ 0.69620253  0.79545455  0.1884058   0.08      ]\n",
+      " [ 0.64556962  0.56818182  0.43478261  0.44      ]\n",
+      " [ 0.63291139  0.79545455  0.1884058   0.12      ]\n",
+      " [ 0.86075949  0.63636364  0.69565217  0.56      ]\n",
+      " [ 0.62025316  0.70454545  0.2173913   0.08      ]\n",
+      " [ 0.73417722  0.90909091  0.17391304  0.08      ]]\n",
+      "Epoch 0 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 1 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 2 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 3 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 4 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 5 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 6 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 7 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 8 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 9 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 10 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 11 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 12 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 13 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 14 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 15 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 16 : 13 / 45 - 28.89 % acc\n",
+      "Epoch 17 : 26 / 45 - 57.78 % acc\n",
+      "Epoch 18 : 27 / 45 - 60.00 % acc\n",
+      "Epoch 19 : 32 / 45 - 71.11 % acc\n",
+      "Epoch 20 : 27 / 45 - 60.00 % acc\n",
+      "Epoch 21 : 28 / 45 - 62.22 % acc\n",
+      "Epoch 22 : 37 / 45 - 82.22 % acc\n",
+      "Epoch 23 : 35 / 45 - 77.78 % acc\n",
+      "Epoch 24 : 32 / 45 - 71.11 % acc\n",
+      "Epoch 25 : 42 / 45 - 93.33 % acc\n",
+      "Epoch 26 : 41 / 45 - 91.11 % acc\n",
+      "Epoch 27 : 36 / 45 - 80.00 % acc\n",
+      "Epoch 28 : 31 / 45 - 68.89 % acc\n",
+      "Epoch 29 : 31 / 45 - 68.89 % acc\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABFsAAAPZCAYAAADOfyVVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nALjoJ2//qvRPFxvxAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f44fc4a2ed0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "iris_data = sns.load_dataset(\"iris\")\n",
+    "\n",
+    "# randomly shuffle data\n",
+    "iris_data = shuffle(iris_data)\n",
+    "\n",
+    "# print first 5 data points\n",
+    "print iris_data[:5]\n",
+    "\n",
+    "\n",
+    "# create pairplot of iris data\n",
+    "g = sns.pairplot(iris_data, hue=\"species\")\n",
+    "\n",
+    "# convert iris data to numpy format\n",
+    "iris_array = iris_data.as_matrix()\n",
+    "\n",
+    "# split data into feature and target sets\n",
+    "X = iris_array[:, :4].astype(float)\n",
+    "y = iris_array[:, -1]\n",
+    "\n",
+    "\n",
+    "# normalize the data per feature by dividing by the maximum value in each column\n",
+    "X = X / X.max(axis=0)\n",
+    "print X\n",
+    "\n",
+    "# convert the textual category data to integer using numpy's unique() function\n",
+    "_, y = np.unique(y, return_inverse=True)\n",
+    "\n",
+    "# convert the list of targets to a vertical matrix with the dimensions [1 x number of samples]\n",
+    "# this is necessary for later computation\n",
+    "y = y.reshape(-1,1)\n",
+    "\n",
+    "# combine feature and target data into a new python array\n",
+    "data = []\n",
+    "for i in range(X.shape[0]):\n",
+    "    data.append(tuple([X[i].reshape(-1,1), y[i][0]]))\n",
+    "\n",
+    "# split data into training and test sets\n",
+    "trainingSplit = int(.7 * len(data))\n",
+    "training_data = data[:trainingSplit]\n",
+    "test_data = data[trainingSplit:]\n",
+    "\n",
+    "# create an instance of the one-hot encoding function from the sci-kit learn library\n",
+    "enc = OneHotEncoder()\n",
+    "\n",
+    "# use the function to figure out how many categories exist in the data\n",
+    "enc.fit(y)\n",
+    "\n",
+    "# convert only the target data in the training set to one-hot encoding\n",
+    "training_data = [[_x, enc.transform(_y.reshape(-1,1)).toarray().reshape(-1,1)] for _x, _y in training_data]\n",
+    "\n",
+    "# define the network\n",
+    "net = Network([4, 32, 3])\n",
+    "\n",
+    "# train the network using SGD, and output the results\n",
+    "results = net.SGD(training_data, 30, 10, 0.2, test_data=test_data)\n",
+    "\n",
+    "# visualize the results\n",
+    "plt.plot(results)\n",
+    "plt.ylabel('accuracy (%)')\n",
+    "plt.ylim([0,100.0])\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[[  1.33900000e+01   1.77000000e+00   2.62000000e+00 ...,   9.20000000e-01\n",
+      "    3.22000000e+00   1.19500000e+03]\n",
+      " [  1.37300000e+01   4.36000000e+00   2.26000000e+00 ...,   7.80000000e-01\n",
+      "    1.75000000e+00   5.20000000e+02]\n",
+      " [  1.41000000e+01   2.16000000e+00   2.30000000e+00 ...,   1.25000000e+00\n",
+      "    3.17000000e+00   1.51000000e+03]\n",
+      " ..., \n",
+      " [  1.28100000e+01   2.31000000e+00   2.40000000e+00 ...,   6.60000000e-01\n",
+      "    1.36000000e+00   5.60000000e+02]\n",
+      " [  1.17600000e+01   2.68000000e+00   2.92000000e+00 ...,   1.23000000e+00\n",
+      "    2.50000000e+00   6.07000000e+02]\n",
+      " [  1.38800000e+01   5.04000000e+00   2.23000000e+00 ...,   5.80000000e-01\n",
+      "    1.33000000e+00   4.15000000e+02]]\n",
+      "[ 0.  2.  0.  1.  0.  1.  0.  0.  0.  2.  2.  2.  0.  2.  1.  2.  1.  2.\n",
+      "  1.  1.  1.  2.  0.  1.  1.  0.  2.  1.  2.  1.  1.  1.  1.  1.  1.  1.\n",
+      "  1.  2.  0.  1.  1.  0.  0.  0.  2.  2.  0.  0.  0.  1.  1.  0.  2.  2.\n",
+      "  0.  1.  2.  1.  1.  2.  0.  1.  2.  1.  2.  2.  0.  0.  1.  0.  1.  0.\n",
+      "  0.  2.  1.  0.  2.  1.  1.  0.  1.  1.  0.  2.  1.  0.  0.  1.  2.  0.\n",
+      "  0.  0.  0.  1.  2.  1.  1.  0.  1.  0.  2.  2.  1.  0.  0.  2.  1.  1.\n",
+      "  0.  1.  2.  1.  2.  1.  1.  1.  2.  1.  0.  1.  2.  1.  2.  0.  1.  1.\n",
+      "  2.  0.  1.  1.  0.  1.  2.  0.  2.  1.  1.  0.  0.  0.  1.  0.  1.  1.\n",
+      "  1.  1.  0.  2.  2.  2.  0.  1.  2.  1.  2.  2.  0.  0.  0.  0.  0.  1.\n",
+      "  1.  0.  2.  0.  1.  2.  1.  2.  0.  0.  0.  2.  1.  2.  1.  2.]\n",
+      "Epoch 0 : 24 / 54 - 44.44 % acc\n",
+      "Epoch 1 : 29 / 54 - 53.70 % acc\n",
+      "Epoch 2 : 21 / 54 - 38.89 % acc\n",
+      "Epoch 3 : 35 / 54 - 64.81 % acc\n",
+      "Epoch 4 : 48 / 54 - 88.89 % acc\n",
+      "Epoch 5 : 42 / 54 - 77.78 % acc\n",
+      "Epoch 6 : 50 / 54 - 92.59 % acc\n",
+      "Epoch 7 : 53 / 54 - 98.15 % acc\n",
+      "Epoch 8 : 50 / 54 - 92.59 % acc\n",
+      "Epoch 9 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 10 : 40 / 54 - 74.07 % acc\n",
+      "Epoch 11 : 52 / 54 - 96.30 % acc\n",
+      "Epoch 12 : 49 / 54 - 90.74 % acc\n",
+      "Epoch 13 : 50 / 54 - 92.59 % acc\n",
+      "Epoch 14 : 52 / 54 - 96.30 % acc\n",
+      "Epoch 15 : 49 / 54 - 90.74 % acc\n",
+      "Epoch 16 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 17 : 49 / 54 - 90.74 % acc\n",
+      "Epoch 18 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 19 : 52 / 54 - 96.30 % acc\n",
+      "Epoch 20 : 52 / 54 - 96.30 % acc\n",
+      "Epoch 21 : 48 / 54 - 88.89 % acc\n",
+      "Epoch 22 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 23 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 24 : 49 / 54 - 90.74 % acc\n",
+      "Epoch 25 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 26 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 27 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 28 : 51 / 54 - 94.44 % acc\n",
+      "Epoch 29 : 51 / 54 - 94.44 % acc\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAssAAAHlCAYAAAADRrEAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f44ca791050>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "wine_data = np.loadtxt(open(\"./data/wine.csv\",\"rb\"),delimiter=\",\")\n",
+    "\n",
+    "wine_data = shuffle(wine_data)\n",
+    "# g = sns.pairplot(wine_data, hue=\"species\")\n",
+    "\n",
+    "X = wine_data[:,1:]\n",
+    "y = wine_data[:, 0] - 1\n",
+    "\n",
+    "print X\n",
+    "print y\n",
+    "\n",
+    "X = X / X.max(axis=0)\n",
+    "y = y.reshape(-1,1)\n",
+    "\n",
+    "data = []\n",
+    "for i in range(X.shape[0]):\n",
+    "    data.append(tuple([X[i].reshape(-1,1), y[i][0]]))\n",
+    "\n",
+    "# split data into training and test sets\n",
+    "trainingSplit = int(.7 * len(data))\n",
+    "training_data = data[:trainingSplit]\n",
+    "test_data = data[trainingSplit:]\n",
+    "\n",
+    "# create an instance of the one-hot encoding function from the sci-kit learn library\n",
+    "enc = OneHotEncoder()\n",
+    "\n",
+    "# use the function to figure out how many categories exist in the data\n",
+    "enc.fit(y)\n",
+    "\n",
+    "# convert only the target data in the training set to one-hot encoding\n",
+    "training_data = [[_x, enc.transform(_y.reshape(-1,1)).toarray().reshape(-1,1)] for _x, _y in training_data]\n",
+    "\n",
+    "# define the network\n",
+    "net = Network([13, 50, 3])\n",
+    "\n",
+    "# train the network using SGD, and output the results\n",
+    "results = net.SGD(training_data, 30, 10, 1.0, test_data=test_data)\n",
+    "\n",
+    "# visualize the results\n",
+    "plt.plot(results)\n",
+    "plt.ylabel('accuracy (%)')\n",
+    "plt.ylim([0,100.0])\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-4/4.2.ipynb b/notebooks/week-4/4.2.ipynb
new file mode 100644
index 0000000..8769d5d
--- /dev/null
+++ b/notebooks/week-4/4.2.ipynb
@@ -0,0 +1,404 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "%matplotlib inline\n",
+    "import math\n",
+    "import random\n",
+    "\n",
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "from sklearn.datasets import load_boston\n",
+    "\n",
+    "'''Since this is a classification problem, we will need to \n",
+    "represent our targets as one-hot encoding vectors (see previous lab).\n",
+    "To do this we will use scikit-learn's OneHotEncoder module \n",
+    "which we import here'''\n",
+    "from sklearn.preprocessing import OneHotEncoder\n",
+    "\n",
+    "import numpy as np\n",
+    "import tensorflow as tf\n",
+    "\n",
+    "sns.set(style=\"ticks\", color_codes=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "dataset = load_boston()\n",
+    "houses = pd.DataFrame(dataset.data, columns=dataset.feature_names)\n",
+    "averageValue = np.mean(dataset.target)\n",
+    "houses['target'] = dataset.target.astype(int)\n",
+    "l_target = len(dataset.target)\n",
+    "\n",
+    "# loop over target values and make binary based on average value\n",
+    "j = 0\n",
+    "while j < l_target:\n",
+    "     if dataset.target[j] > averageValue:\n",
+    "         houses.target.set_value(j, 1)\n",
+    "     else:\n",
+    "         houses.target.set_value(j, 0)\n",
+    "     j += 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 75,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1 <-- should be 1\n",
+      "0 <-- should be 0\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "'''check your work'''\n",
+    "# print houses['target']\n",
+    "# print averageValue\n",
+    "print np.max(houses['target']), \"<-- should be 1\"\n",
+    "print np.min(houses['target']), \"<-- should be 0\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Training set', (354, 13), (354, 2))\n",
+      "('Test set', (152, 13), (152, 2))\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "houses_array = houses.as_matrix().astype(float)\n",
+    "np.random.shuffle(houses_array)\n",
+    "\n",
+    "X = houses_array[:, :-1]\n",
+    "y = houses_array[:, -1]\n",
+    "\n",
+    "\n",
+    "y = y.reshape(-1,1)\n",
+    "\n",
+    "enc = OneHotEncoder()\n",
+    "\n",
+    "enc.fit(y)\n",
+    "\n",
+    "y = enc.transform(y).toarray()\n",
+    "X = X / X.max(axis=0)\n",
+    "\n",
+    "trainingSplit = int(.7 * houses_array.shape[0])\n",
+    "X_train = X[:trainingSplit]\n",
+    "y_train = y[:trainingSplit]\n",
+    "X_test = X[trainingSplit:]\n",
+    "y_test = y[trainingSplit:]\n",
+    "\n",
+    "print('Training set', X_train.shape, y_train.shape)\n",
+    "print('Test set', X_test.shape, y_test.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 77,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2 <-- should be 2\n",
+      "2 <-- should be 2\n",
+      "[ 0.  1.] <-- should be either [0. 1.] or [1. 0.]\n"
+     ]
+    }
+   ],
+   "source": [
+    "'''check your work'''\n",
+    "print y_train.shape[1], \"<-- should be 2\"\n",
+    "print y_test.shape[1], \"<-- should be 2\"\n",
+    "print y_train[0], \"<-- should be either [0. 1.] or [1. 0.]\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# helper variables\n",
+    "num_samples = X_train.shape[0]\n",
+    "num_features = X_train.shape[1]\n",
+    "num_outputs = y_train.shape[1]\n",
+    "\n",
+    "# Hyper-parameters\n",
+    "batch_size = 15\n",
+    "num_hidden_1 = 35\n",
+    "num_hidden_2 =  35\n",
+    "learning_rate =  0.08\n",
+    "training_epochs =  400\n",
+    "dropout_keep_prob = 1 # 0.5 # set to no dropout by default\n",
+    "\n",
+    "# variable to control the resolution at which the training results are stored\n",
+    "display_step = 10"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 79,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def accuracy(predictions, targets):\n",
+    "      \n",
+    "    accuracy = (np.sum((np.argmax(predictions, 1) == np.argmax(targets,1)  ))/float(np.argmax(predictions, 1).shape[0]))*100.0\n",
+    "    \n",
+    "    return accuracy\n",
+    "\n",
+    "def weight_variable(shape):\n",
+    "    initial = tf.truncated_normal(shape, stddev=0.1)\n",
+    "    return tf.Variable(initial)\n",
+    "\n",
+    "def bias_variable(shape):\n",
+    "    initial = tf.constant(0.1, shape=shape)\n",
+    "    return tf.Variable(initial)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 80,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def accuracy(predictions, targets):\n",
+    "      \n",
+    "    accuracy = (np.sum((np.argmax(predictions, 1) == np.argmax(targets,1)  ))/float(np.argmax(predictions, 1).shape[0]))*100.0\n",
+    "    \n",
+    "    return accuracy\n",
+    "\n",
+    "def weight_variable(shape):\n",
+    "    initial = tf.truncated_normal(shape, stddev=0.1)\n",
+    "    return tf.Variable(initial)\n",
+    "\n",
+    "def bias_variable(shape):\n",
+    "    initial = tf.constant(0.1, shape=shape)\n",
+    "    return tf.Variable(initial)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 81,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "graph = tf.Graph()\n",
+    "\n",
+    "with graph.as_default():\n",
+    "    \n",
+    "    x = tf.placeholder(tf.float32, shape=(None, num_features))\n",
+    "    _y = tf.placeholder(tf.float32, shape=(None))\n",
+    "    \n",
+    "    keep_prob = tf.placeholder(tf.float32)\n",
+    "    \n",
+    "    tf_X_test = tf.constant(X_test, dtype=tf.float32)\n",
+    "    tf_X_train = tf.constant(X_train, dtype=tf.float32)\n",
+    "    \n",
+    "    \n",
+    "    W_fc1 = weight_variable([num_features, num_hidden_1])\n",
+    "    b_fc1 = bias_variable([num_hidden_1])\n",
+    "    \n",
+    "    W_fc2 = weight_variable([num_hidden_1, num_hidden_2])\n",
+    "    b_fc2 = bias_variable([num_hidden_2])\n",
+    "    \n",
+    "    W_fc3 = weight_variable([num_hidden_2, num_outputs])\n",
+    "    b_fc3 = bias_variable([num_outputs])\n",
+    "    \n",
+    "    \n",
+    "    def model(data, keep):\n",
+    "        \n",
+    "        fc1 = tf.nn.sigmoid(tf.matmul(data, W_fc1) + b_fc1)\n",
+    "        fc1_drop = tf.nn.dropout(fc1, keep)\n",
+    "        \n",
+    "        fc2 = tf.nn.sigmoid(tf.matmul(fc1_drop, W_fc2) + b_fc2)\n",
+    "        fc2_drop = tf.nn.dropout(fc2, keep)\n",
+    "        \n",
+    "        fc3 = tf.matmul(fc2_drop, W_fc3) + b_fc3\n",
+    "        \n",
+    "        return fc3\n",
+    "    \n",
+    "    '''for our loss function we still want to get the raw outputs \n",
+    "    of the model, but since it no longer represents the actual prediction \n",
+    "    we rename the variable to ‘output’'''\n",
+    "    output = model(x, keep_prob)\n",
+    "    \n",
+    "    # WHEN WE CALCULATE THE PREDICTIONS, WE NEED TO WRAP EACH OUTPUT IN A\n",
+    "    # tf.nn.softmax() FUNCTION. THE FIRST ONE HAS BEEN DONE FOR YOU:\n",
+    "    prediction = tf.nn.softmax(output)\n",
+    "    test_prediction = model(tf_X_test, 1.0)\n",
+    "    train_prediction = model(tf_X_train, 1.0)\n",
+    "    \n",
+    "    '''finally, we replace our previous MSE cost function with the\n",
+    "    cross-entropy function included in Tensorflow. This function takes in the\n",
+    "    raw output of the network and calculates the average loss with the target'''\n",
+    "    loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(output, _y))\n",
+    "    optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss)\n",
+    "    \n",
+    "    saver = tf.train.Saver()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 82,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Initialized\n",
+      "Model saved in file: model_houses_classification.ckpt\n"
+     ]
+    }
+   ],
+   "source": [
+    "results = []\n",
+    "\n",
+    "with tf.Session(graph=graph) as session:\n",
+    "    \n",
+    "    tf.initialize_all_variables().run()\n",
+    "    print('Initialized')\n",
+    "\n",
+    "    for epoch in range(training_epochs):\n",
+    "        \n",
+    "        indexes = range(num_samples)\n",
+    "        random.shuffle(indexes)\n",
+    "        \n",
+    "        for step in range(int(math.floor(num_samples/float(batch_size)))):\n",
+    "            offset = step * batch_size\n",
+    "            \n",
+    "            batch_data = X_train[indexes[offset:(offset + batch_size)]]\n",
+    "            batch_labels = y_train[indexes[offset:(offset + batch_size)]]\n",
+    "\n",
+    "            feed_dict = {x : batch_data, _y : batch_labels, keep_prob: dropout_keep_prob}\n",
+    "            \n",
+    "            _, l, p = session.run([optimizer, loss, prediction], feed_dict=feed_dict)\n",
+    "\n",
+    "        if (epoch % display_step == 0):\n",
+    "            batch_acc = accuracy(p, batch_labels)\n",
+    "            train_acc =  accuracy(train_prediction.eval(session=session), y_train)\n",
+    "            test_acc =  accuracy(test_prediction.eval(session=session), y_test)\n",
+    "            results.append([epoch, batch_acc, train_acc, test_acc])\n",
+    "\n",
+    "    save_path = saver.save(session, \"model_houses_classification.ckpt\")\n",
+    "    print(\"Model saved in file: %s\" % save_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 83,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Maximum test accuracy: 90.79%\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAAGUCAYAAABELNFLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAQYGjUAAAAACDA/D/jnxysCRYkwQAAAABJRU5ErkJggg==\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f576452be50>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "df = pd.DataFrame(data=results, columns = [\"epoch\", \"batch_acc\", \"train_acc\", \"test_acc\"])\n",
+    "df.set_index(\"epoch\", drop=True, inplace=True)\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(10, 4))\n",
+    " \n",
+    "ax.plot(df)\n",
+    "ax.set(xlabel='Epoch',\n",
+    "       ylabel='Error',\n",
+    "       title='Training result')\n",
+    " \n",
+    "ax.legend(df.columns, loc=1)\n",
+    "\n",
+    "print \"Maximum test accuracy: %.2f%%\" % np.max(df[\"test_acc\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-4/labbb 4.1.ipynb b/notebooks/week-4/labbb 4.1.ipynb
new file mode 100644
index 0000000..120270a
--- /dev/null
+++ b/notebooks/week-4/labbb 4.1.ipynb	
@@ -0,0 +1,554 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "%matplotlib inline\n",
+    "import math\n",
+    "import random\n",
+    "\n",
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "from sklearn.datasets import load_boston\n",
+    "\n",
+    "import numpy as np\n",
+    "import tensorflow as tf\n",
+    "\n",
+    "sns.set(style=\"ticks\", color_codes=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "      CRIM    ZN  INDUS  CHAS    NOX     RM   AGE     DIS  RAD    TAX  \\\n",
+      "0  0.00632  18.0   2.31   0.0  0.538  6.575  65.2  4.0900  1.0  296.0   \n",
+      "1  0.02731   0.0   7.07   0.0  0.469  6.421  78.9  4.9671  2.0  242.0   \n",
+      "2  0.02729   0.0   7.07   0.0  0.469  7.185  61.1  4.9671  2.0  242.0   \n",
+      "3  0.03237   0.0   2.18   0.0  0.458  6.998  45.8  6.0622  3.0  222.0   \n",
+      "4  0.06905   0.0   2.18   0.0  0.458  7.147  54.2  6.0622  3.0  222.0   \n",
+      "\n",
+      "   PTRATIO       B  LSTAT  target  \n",
+      "0     15.3  396.90   4.98    24.0  \n",
+      "1     17.8  396.90   9.14    21.6  \n",
+      "2     17.8  392.83   4.03    34.7  \n",
+      "3     18.7  394.63   2.94    33.4  \n",
+      "4     18.7  396.90   5.33    36.2  \n"
+     ]
+    }
+   ],
+   "source": [
+    "#load data from scikit-learn library\n",
+    "dataset = load_boston()\n",
+    "\n",
+    "#load data as DataFrame\n",
+    "houses = pd.DataFrame(dataset.data, columns=dataset.feature_names)\n",
+    "#add target data to DataFrame\n",
+    "houses['target'] = dataset.target\n",
+    "\n",
+    "#print first 5 entries of data\n",
+    "print houses.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Boston House Prices dataset\n",
+      "===========================\n",
+      "\n",
+      "Notes\n",
+      "------\n",
+      "Data Set Characteristics:  \n",
+      "\n",
+      "    :Number of Instances: 506 \n",
+      "\n",
+      "    :Number of Attributes: 13 numeric/categorical predictive\n",
+      "    \n",
+      "    :Median Value (attribute 14) is usually the target\n",
+      "\n",
+      "    :Attribute Information (in order):\n",
+      "        - CRIM     per capita crime rate by town\n",
+      "        - ZN       proportion of residential land zoned for lots over 25,000 sq.ft.\n",
+      "        - INDUS    proportion of non-retail business acres per town\n",
+      "        - CHAS     Charles River dummy variable (= 1 if tract bounds river; 0 otherwise)\n",
+      "        - NOX      nitric oxides concentration (parts per 10 million)\n",
+      "        - RM       average number of rooms per dwelling\n",
+      "        - AGE      proportion of owner-occupied units built prior to 1940\n",
+      "        - DIS      weighted distances to five Boston employment centres\n",
+      "        - RAD      index of accessibility to radial highways\n",
+      "        - TAX      full-value property-tax rate per $10,000\n",
+      "        - PTRATIO  pupil-teacher ratio by town\n",
+      "        - B        1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town\n",
+      "        - LSTAT    % lower status of the population\n",
+      "        - MEDV     Median value of owner-occupied homes in $1000's\n",
+      "\n",
+      "    :Missing Attribute Values: None\n",
+      "\n",
+      "    :Creator: Harrison, D. and Rubinfeld, D.L.\n",
+      "\n",
+      "This is a copy of UCI ML housing dataset.\n",
+      "http://archive.ics.uci.edu/ml/datasets/Housing\n",
+      "\n",
+      "\n",
+      "This dataset was taken from the StatLib library which is maintained at Carnegie Mellon University.\n",
+      "\n",
+      "The Boston house-price data of Harrison, D. and Rubinfeld, D.L. 'Hedonic\n",
+      "prices and the demand for clean air', J. Environ. Economics & Management,\n",
+      "vol.5, 81-102, 1978.   Used in Belsley, Kuh & Welsch, 'Regression diagnostics\n",
+      "...', Wiley, 1980.   N.B. Various transformations are used in the table on\n",
+      "pages 244-261 of the latter.\n",
+      "\n",
+      "The Boston house-price data has been used in many machine learning papers that address regression\n",
+      "problems.   \n",
+      "     \n",
+      "**References**\n",
+      "\n",
+      "   - Belsley, Kuh & Welsch, 'Regression diagnostics: Identifying Influential Data and Sources of Collinearity', Wiley, 1980. 244-261.\n",
+      "   - Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning. In Proceedings on the Tenth International Conference of Machine Learning, 236-243, University of Massachusetts, Amherst. Morgan Kaufmann.\n",
+      "   - many more! (see http://archive.ics.uci.edu/ml/datasets/Housing)\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "print dataset['DESCR']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsEAAAJTCAYAAADpH8s8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f7b666a0210>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Create a datset of correlations between house features\n",
+    "corrmat = houses.corr()\n",
+    "\n",
+    "# Set up the matplotlib figure\n",
+    "f, ax = plt.subplots(figsize=(9, 6))\n",
+    "\n",
+    "# Draw the heatmap using seaborn\n",
+    "sns.set_context(\"notebook\", font_scale=0.7, rc={\"lines.linewidth\": 1.5})\n",
+    "sns.heatmap(corrmat, annot=True, square=True)\n",
+    "f.tight_layout()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<seaborn.axisgrid.JointGrid at 0x7f7b63a6a7d0>"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkYAAAJRCAYAAACgDnhPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCAAAIEUwAgAASP0/SIcBFuu07esAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f7b63cc4750>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAJRCAYAAABY7oO4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhiYiIiIiE/4/uaZc8TIBg+gAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f7b63a6a650>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.jointplot(houses['target'], houses['RM'], kind='hex')\n",
+    "sns.jointplot(houses['target'], houses['LSTAT'], kind='hex')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Training set', (354, 13), (354,))\n",
+      "('Test set', (152, 13), (152,))\n"
+     ]
+    }
+   ],
+   "source": [
+    "# convert housing data to numpy format\n",
+    "houses_array = houses.as_matrix().astype(float)\n",
+    "\n",
+    "# split data into feature and target sets\n",
+    "X = houses_array[:, :-1]\n",
+    "y = houses_array[:, -1]\n",
+    "\n",
+    "# normalize the data per feature by dividing by the maximum value in each column\n",
+    "X = X / X.max(axis=0)\n",
+    "\n",
+    "# split data into training and test sets\n",
+    "trainingSplit = int(.7 * houses_array.shape[0])\n",
+    "\n",
+    "X_train = X[:trainingSplit]\n",
+    "y_train = y[:trainingSplit]\n",
+    "X_test = X[trainingSplit:]\n",
+    "y_test = y[trainingSplit:]\n",
+    "\n",
+    "print('Training set', X_train.shape, y_train.shape)\n",
+    "print('Test set', X_test.shape, y_test.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# helper variables\n",
+    "num_samples = X_train.shape[0]\n",
+    "num_features = X_train.shape[1]\n",
+    "num_outputs = 1\n",
+    "\n",
+    "# Hyper-parameters\n",
+    "batch_size = 50\n",
+    "num_hidden_1 = 30\n",
+    "num_hidden_2 = 30\n",
+    "learning_rate = 0.0001\n",
+    "training_epochs = 200\n",
+    "dropout_keep_prob = 0.2 # this seems to be a typical value for dropout???\n",
+    "\n",
+    "# variable to control the resolution at which the training results are stored\n",
+    "display_step = 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def accuracy(predictions, targets):\n",
+    "    error = np.absolute(predictions.reshape(-1) - targets)\n",
+    "    return np.mean(error)\n",
+    "\n",
+    "def weight_variable(shape):\n",
+    "    initial = tf.truncated_normal(shape, stddev=0.1) # creates trunc gauss dist of weights\n",
+    "    return tf.Variable(initial)\n",
+    "\n",
+    "def bias_variable(shape):\n",
+    "    initial = tf.constant(0.1, shape=shape) # starts with a bias of 0.1\n",
+    "    return tf.Variable(initial)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "'''First we create a variable to store our graph'''\n",
+    "graph = tf.Graph()\n",
+    "\n",
+    "'''Next we build our neural network within this graph variable'''\n",
+    "with graph.as_default():\n",
+    "    \n",
+    "    '''Our training data will come in as x feature data and \n",
+    "    y target data. We need to create tensorflow placeholders \n",
+    "    to capture this data as it comes in'''\n",
+    "    \n",
+    "    x = tf.placeholder(tf.float32, shape=(None, num_features))\n",
+    "    _y = tf.placeholder(tf.float32, shape=(None))\n",
+    "    \n",
+    "    '''Another placeholder stores the hyperparameter \n",
+    "    that controls dropout'''\n",
+    "    \n",
+    "    keep_prob = tf.placeholder(tf.float32)\n",
+    "    \n",
+    "    '''Finally, we convert the test and train feature data sets \n",
+    "    to tensorflow constants so we can use them to generate \n",
+    "    predictions on both data sets'''\n",
+    "    \n",
+    "    tf_X_test = tf.constant(X_test, dtype=tf.float32)\n",
+    "    tf_X_train = tf.constant(X_train, dtype=tf.float32)\n",
+    "    \n",
+    "    '''Next we create the parameter variables for the model.\n",
+    "    Each layer of the neural network needs it's own weight \n",
+    "    and bias variables which will be tuned during training.\n",
+    "    The sizes of the parameter variables are determined by \n",
+    "    the number of neurons in each layer.'''\n",
+    "    \n",
+    "    W_fc1 = weight_variable([num_features, num_hidden_1])\n",
+    "    b_fc1 = bias_variable([num_hidden_1])\n",
+    "    \n",
+    "    W_fc2 = weight_variable([num_hidden_1, num_hidden_2])\n",
+    "    b_fc2 = bias_variable([num_hidden_2])\n",
+    "    \n",
+    "    W_fc3 = weight_variable([num_hidden_2, num_outputs])\n",
+    "    b_fc3 = bias_variable([num_outputs])\n",
+    "    \n",
+    "    '''Next, we define the forward computation of the model.\n",
+    "    We do this by defining a function model() which takes in \n",
+    "    a set of input data, and performs computations through \n",
+    "    the network until it generates the output.'''\n",
+    "    \n",
+    "    def model(data, keep):\n",
+    "        \n",
+    "        # computing first hidden layer from input, using relu activation function\n",
+    "        fc1 = tf.nn.sigmoid(tf.matmul(data, W_fc1) + b_fc1)\n",
+    "        # adding dropout to first hidden layer\n",
+    "        fc1_drop = tf.nn.dropout(fc1, keep)\n",
+    "        \n",
+    "        # computing second hidden layer from first hidden layer, using relu activation function\n",
+    "        fc2 = tf.nn.sigmoid(tf.matmul(fc1_drop, W_fc2) + b_fc2)\n",
+    "        # adding dropout to second hidden layer\n",
+    "        fc2_drop = tf.nn.dropout(fc2, keep)\n",
+    "        \n",
+    "        # computing output layer from second hidden layer\n",
+    "        # the output is a single neuron which is directly interpreted as the prediction of the target value\n",
+    "        fc3 = tf.matmul(fc2_drop, W_fc3) + b_fc3\n",
+    "        \n",
+    "        # the output is returned from the function\n",
+    "        return fc3\n",
+    "    \n",
+    "    '''Next we define a few calls to the model() function which \n",
+    "    will return predictions for the current batch input data (x),\n",
+    "    as well as the entire test and train feature set'''\n",
+    "    \n",
+    "    prediction = model(x, keep_prob)\n",
+    "    test_prediction = model(tf_X_test, 1.0)\n",
+    "    train_prediction = model(tf_X_train, 1.0)\n",
+    "    \n",
+    "    '''Finally, we define the loss and optimization functions \n",
+    "    which control how the model is trained.\n",
+    "    \n",
+    "    For the loss we will use the basic mean square error (MSE) function,\n",
+    "    which tries to minimize the MSE between the predicted values and the \n",
+    "    real values (_y) of the input dataset.\n",
+    "    \n",
+    "    For the optimization function we will use basic Gradient Descent (SGD)\n",
+    "    which will minimize the loss using the specified learning rate.'''\n",
+    "    \n",
+    "    loss = tf.reduce_mean(tf.square(tf.sub(prediction, _y)))\n",
+    "    optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss)\n",
+    "    \n",
+    "    '''We also create a saver variable which will allow us to \n",
+    "    save our trained model for later use'''\n",
+    "    \n",
+    "    saver = tf.train.Saver()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Initialized\n",
+      "Model saved in file: model_houses.ckpt\n"
+     ]
+    }
+   ],
+   "source": [
+    "# create an array to store the results of the optimization at each epoch\n",
+    "results = []\n",
+    "\n",
+    "'''First we open a session of Tensorflow using our graph as the base. \n",
+    "While this session is active all the parameter values will be stored, \n",
+    "and each step of training will be using the same model.'''\n",
+    "with tf.Session(graph=graph) as session:\n",
+    "    \n",
+    "    '''After we start a new session we first need to\n",
+    "    initialize the values of all the variables.'''\n",
+    "    tf.initialize_all_variables().run() # I don't understand what variables are being initialized...\n",
+    "    print('Initialized')\n",
+    "\n",
+    "    '''Now we iterate through each training epoch based on the hyper-parameter set above.\n",
+    "    Each epoch represents a single pass through all the training data.\n",
+    "    The total number of training steps is determined by the number of epochs and \n",
+    "    the size of mini-batches relative to the size of the entire training set.'''\n",
+    "    for epoch in range(training_epochs):\n",
+    "        \n",
+    "        '''At the beginning of each epoch, we create a set of shuffled indexes \n",
+    "        so that we are using the training data in a different order each time'''\n",
+    "        indexes = range(num_samples)\n",
+    "        random.shuffle(indexes)\n",
+    "        \n",
+    "        '''Next we step through each mini-batch in the training set'''\n",
+    "        for step in range(int(math.floor(num_samples/float(batch_size)))):\n",
+    "            offset = step * batch_size\n",
+    "            \n",
+    "            '''We subset the feature and target training sets to create each mini-batch'''\n",
+    "            batch_data = X_train[indexes[offset:(offset + batch_size)]]\n",
+    "            batch_labels = y_train[indexes[offset:(offset + batch_size)]]\n",
+    "\n",
+    "            '''Then, we create a 'feed dictionary' that will feed this data, \n",
+    "            along with any other hyper-parameters such as the dropout probability,\n",
+    "            to the model'''\n",
+    "            feed_dict = {x : batch_data, _y : batch_labels, keep_prob: dropout_keep_prob}\n",
+    "            \n",
+    "            '''Finally, we call the session's run() function, which will feed in \n",
+    "            the current training data, and execute portions of the graph as necessary \n",
+    "            to return the data we ask for.\n",
+    "            \n",
+    "            The first argument of the run() function is a list specifying the \n",
+    "            model variables we want it to compute and return from the function. \n",
+    "            The most important is 'optimizer' which triggers all calculations necessary \n",
+    "            to perform one training step. We also include 'loss' and 'prediction' \n",
+    "            because we want these as ouputs from the function so we can keep \n",
+    "            track of the training process.\n",
+    "            \n",
+    "            The second argument specifies the feed dictionary that contains \n",
+    "            all the data we want to pass into the model at each training step.'''\n",
+    "            _, l, p = session.run([optimizer, loss, prediction], feed_dict=feed_dict)\n",
+    "\n",
+    "        '''At the end of each epoch, we will calculate the error of predictions \n",
+    "        on the full training and test data set. We will then store the epoch number, \n",
+    "        along with the mini-batch, training, and test accuracies to the 'results' array \n",
+    "        so we can visualize the training process later. How often we save the data to \n",
+    "        this array is specified by the display_step variable created above''' \n",
+    "        if (epoch % display_step == 0):\n",
+    "            batch_acc = accuracy(p, batch_labels)\n",
+    "            train_acc =  accuracy(train_prediction.eval(session=session), y_train)\n",
+    "            test_acc =  accuracy(test_prediction.eval(session=session), y_test)\n",
+    "            results.append([epoch, batch_acc, train_acc, test_acc])\n",
+    "\n",
+    "    '''Once training is complete, we will save the trained model so that we can use it later'''\n",
+    "    save_path = saver.save(session, \"model_houses.ckpt\")\n",
+    "    print(\"Model saved in file: %s\" % save_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Minimum test loss: 5.53774121561\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAAGHCAYAAABViAiMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzQAAAABJRU5ErkJggg==\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f7b61efc3d0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "df = pd.DataFrame(data=results, columns = [\"epoch\", \"batch_acc\", \"train_acc\", \"test_acc\"])\n",
+    "df.set_index(\"epoch\", drop=True, inplace=True)\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(10, 4))\n",
+    " \n",
+    "ax.plot(df)\n",
+    "ax.set(xlabel='Epoch',\n",
+    "       ylabel='Error',\n",
+    "       title='Training result')\n",
+    " \n",
+    "ax.legend(df.columns, loc=1)\n",
+    "\n",
+    "print \"Minimum test loss:\", np.min(df[\"test_acc\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-5/5.3.ipynb b/notebooks/week-5/5.3.ipynb
new file mode 100644
index 0000000..f87a80d
--- /dev/null
+++ b/notebooks/week-5/5.3.ipynb
@@ -0,0 +1,407 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Training set', (14000, 64, 64), (14000, 1))\n",
+      "('Test set', (6000, 64, 64), (6000, 1))\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "import pickle\n",
+    "\n",
+    "pickle_file = '-catsdogs.pickle'\n",
+    "\n",
+    "with open(pickle_file, 'rb') as f:\n",
+    "    save = pickle.load(f)\n",
+    "    X_train = save['X_train']\n",
+    "    y_train = save['y_train']\n",
+    "    X_test = save['X_test']\n",
+    "    y_test = save['y_test']\n",
+    "    del save  # hint to help gc free up memory\n",
+    "    print('Training set', X_train.shape, y_train.shape)\n",
+    "    print('Test set', X_test.shape, y_test.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "using ordering: tf\n"
+     ]
+    }
+   ],
+   "source": [
+    "## implement your CNN starting here.\n",
+    "import pickle\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense, Dropout, Activation, Flatten\n",
+    "from keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D\n",
+    "from keras.utils import np_utils\n",
+    "from keras.optimizers import SGD\n",
+    "from keras import backend as K\n",
+    "\n",
+    "K.set_image_dim_ordering('tf')\n",
+    "\n",
+    "print \"using ordering:\", K.image_dim_ordering()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(14000, 64, 64)\n",
+      "64 64\n",
+      "(14000, 64, 64, 1)\n",
+      "(14000, 1)\n"
+     ]
+    }
+   ],
+   "source": [
+    "num_classes = 2\n",
+    "img_rows, img_cols = X_train.shape[1],  X_train.shape[2]\n",
+    "\n",
+    "print X_train.shape\n",
+    "print img_rows, img_cols\n",
+    "\n",
+    "if K.image_dim_ordering()== 'th':\n",
+    "    X_train = X_train.reshape(X_train.shape[0],1, img_rows, img_cols)\n",
+    "    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)\n",
+    "    input_shape = (1, img_rows, img_cols)\n",
+    "else:\n",
+    "    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
+    "    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
+    "    input_shape = (img_rows, img_cols, 1)\n",
+    "\n",
+    "Y_train = np_utils.to_categorical(y_train, num_classes)\n",
+    "Y_test = np_utils.to_categorical(y_test, num_classes)\n",
+    "\n",
+    "print X_train.shape\n",
+    "print y_train.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(64, 64)\n",
+      "(14000, 1)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7f13d211b9d0>"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFiCAYAAAAna2l5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f13d67a1d50>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "from matplotlib.pyplot import imshow\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "img_num = 0\n",
+    "\n",
+    "if K.image_dim_ordering() == 'th':\n",
+    "    img = X_train[img_num][0,:,:]\n",
+    "else:\n",
+    "    img = X_train[img_num][:,:,0]\n",
+    "\n",
+    "print img.shape\n",
+    "print y_train.shape\n",
+    "imshow(img, cmap = plt.get_cmap('gray'), vmin = 0, vmax = 1,  interpolation='nearest')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# model hyperparameters\n",
+    "batch_size = 128\n",
+    "nb_epoch = 30\n",
+    "\n",
+    "# network architecture\n",
+    "patch_size_1 = 3\n",
+    "patch_size_2 = 3\n",
+    "patch_size_3 = 3\n",
+    "\n",
+    "depth_1 = 20\n",
+    "depth_2 = 40\n",
+    "depth_3 = 80\n",
+    "\n",
+    "pool_size = 2\n",
+    "\n",
+    "num_hidden_1 = 1000\n",
+    "num_hidden_2 = 1000\n",
+    "num_hidden_3 = 1000\n",
+    "\n",
+    "dropout = 0.5"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# create new Keras Sequential model\n",
+    "model = Sequential()\n",
+    "\n",
+    "# add first convolutional layer to model and specify it's depth and filter size\n",
+    "# for the first layer we also have to specify the size of each input image\n",
+    "# which we calculated above\n",
+    "model.add(Convolution2D(depth_1, patch_size_1, patch_size_1,\n",
+    "                        border_mode='valid',\n",
+    "                        input_shape=input_shape))\n",
+    "# apply 'relu' activation function for first layer\n",
+    "model.add(Activation('relu'))\n",
+    "# apply max pooling to reduce the size of the image by a factor of 2\n",
+    "model.add(MaxPooling2D(pool_size=(pool_size, pool_size)))\n",
+    "\n",
+    "# repeat these operations for the second convolutional layer\n",
+    "# this time Keras can figure out the input size \n",
+    "# from the previous layer on it's own\n",
+    "model.add(Convolution2D(depth_2, patch_size_2, patch_size_2,\n",
+    "                        border_mode='valid'))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(MaxPooling2D(pool_size=(pool_size, pool_size)))\n",
+    "\n",
+    "model.add(Convolution2D(depth_3, patch_size_3, patch_size_3,border_mode='valid'))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(MaxPooling2D(pool_size=(pool_size, pool_size)))\n",
+    "\n",
+    "# flatten the three-dimensional convolutional layer to a single layer of neurons\n",
+    "model.add(Flatten())\n",
+    "\n",
+    "# add the first fully connected layer, applying 'relu' activation and dropout\n",
+    "model.add(Dense(num_hidden_1))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(Dropout(dropout))\n",
+    "\n",
+    "# add the second fully connected layer\n",
+    "model.add(Dense(num_hidden_2))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(Dropout(dropout))\n",
+    "\n",
+    "# add the third fully connected layer\n",
+    "model.add(Dense(num_hidden_3))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(Dropout(dropout))\n",
+    "\n",
+    "# add the final classification layer with the number of neurons \n",
+    "# matching the number of classes we are trying to learn\n",
+    "model.add(Dense(num_classes))\n",
+    "\n",
+    "# apply the 'softmax' activation to the final layer to convert the output to \n",
+    "# a probability distribution\n",
+    "model.add(Activation('softmax'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Train on 14000 samples, validate on 6000 samples\n",
+      "Epoch 1/30\n",
+      "14000/14000 [==============================] - 96s - loss: 0.6901 - acc: 0.5311 - val_loss: 0.6931 - val_acc: 0.5150\n",
+      "Epoch 2/30\n",
+      "14000/14000 [==============================] - 88s - loss: 0.6736 - acc: 0.5834 - val_loss: 0.7237 - val_acc: 0.5178\n",
+      "Epoch 3/30\n",
+      "14000/14000 [==============================] - 86s - loss: 0.6589 - acc: 0.6099 - val_loss: 0.6373 - val_acc: 0.6477\n",
+      "Epoch 4/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.6446 - acc: 0.6296 - val_loss: 0.6252 - val_acc: 0.6613\n",
+      "Epoch 5/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.6149 - acc: 0.6604 - val_loss: 0.6993 - val_acc: 0.5890\n",
+      "Epoch 6/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.5875 - acc: 0.6909 - val_loss: 0.5573 - val_acc: 0.7145\n",
+      "Epoch 7/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.5572 - acc: 0.7144 - val_loss: 0.6241 - val_acc: 0.6602\n",
+      "Epoch 8/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.5343 - acc: 0.7329 - val_loss: 0.7225 - val_acc: 0.5972\n",
+      "Epoch 9/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.5002 - acc: 0.7589 - val_loss: 0.4825 - val_acc: 0.7657\n",
+      "Epoch 10/30\n",
+      "14000/14000 [==============================] - 86s - loss: 0.4784 - acc: 0.7698 - val_loss: 0.4921 - val_acc: 0.7587\n",
+      "Epoch 11/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.4526 - acc: 0.7886 - val_loss: 0.5201 - val_acc: 0.7560\n",
+      "Epoch 12/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.4352 - acc: 0.7986 - val_loss: 0.4676 - val_acc: 0.7710\n",
+      "Epoch 13/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.4098 - acc: 0.8088 - val_loss: 0.4805 - val_acc: 0.7713\n",
+      "Epoch 14/30\n",
+      "14000/14000 [==============================] - 86s - loss: 0.3910 - acc: 0.8249 - val_loss: 0.5015 - val_acc: 0.7793\n",
+      "Epoch 15/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.3559 - acc: 0.8434 - val_loss: 0.5509 - val_acc: 0.7643\n",
+      "Epoch 16/30\n",
+      "14000/14000 [==============================] - 87s - loss: 0.3346 - acc: 0.8554 - val_loss: 0.4756 - val_acc: 0.7982\n",
+      "Epoch 17/30\n",
+      "14000/14000 [==============================] - 86s - loss: 0.3017 - acc: 0.8712 - val_loss: 0.4689 - val_acc: 0.7945\n",
+      "Epoch 18/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.2720 - acc: 0.8886 - val_loss: 0.4788 - val_acc: 0.8087\n",
+      "Epoch 19/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.2329 - acc: 0.9071 - val_loss: 0.4877 - val_acc: 0.7958\n",
+      "Epoch 20/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.1902 - acc: 0.9225 - val_loss: 0.5330 - val_acc: 0.8042\n",
+      "Epoch 21/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.1552 - acc: 0.9409 - val_loss: 0.6219 - val_acc: 0.7938\n",
+      "Epoch 22/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.1228 - acc: 0.9533 - val_loss: 0.6120 - val_acc: 0.8047\n",
+      "Epoch 23/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0983 - acc: 0.9651 - val_loss: 0.7916 - val_acc: 0.7830\n",
+      "Epoch 24/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0731 - acc: 0.9751 - val_loss: 0.7471 - val_acc: 0.8060\n",
+      "Epoch 25/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0594 - acc: 0.9780 - val_loss: 0.8660 - val_acc: 0.7932\n",
+      "Epoch 26/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0465 - acc: 0.9834 - val_loss: 0.8665 - val_acc: 0.8072\n",
+      "Epoch 27/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0431 - acc: 0.9839 - val_loss: 0.8880 - val_acc: 0.8070\n",
+      "Epoch 28/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0294 - acc: 0.9905 - val_loss: 0.9628 - val_acc: 0.7968\n",
+      "Epoch 29/30\n",
+      "14000/14000 [==============================] - 84s - loss: 0.0303 - acc: 0.9896 - val_loss: 0.9726 - val_acc: 0.8060\n",
+      "Epoch 30/30\n",
+      "14000/14000 [==============================] - 85s - loss: 0.0233 - acc: 0.9914 - val_loss: 1.0685 - val_acc: 0.8045\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<keras.callbacks.History at 0x7f13cce14e50>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.compile(loss='categorical_crossentropy',\n",
+    "              optimizer='adadelta',\n",
+    "              metrics=['accuracy'])\n",
+    "model.fit(X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch,\n",
+    "          verbose=1, validation_data=(X_test, Y_test))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test score: 1.06850525743\n",
+      "Test accuracy: 80.45%\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "score = model.evaluate(X_test, Y_test, verbose=0)\n",
+    "\n",
+    "print 'Test score:', score[0]\n",
+    "print 'Test accuracy: {:.2%}'.format(score[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-5/assignment 5.2.ipynb b/notebooks/week-5/assignment 5.2.ipynb
new file mode 100644
index 0000000..747e244
--- /dev/null
+++ b/notebooks/week-5/assignment 5.2.ipynb	
@@ -0,0 +1,262 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "\n",
+    "%matplotlib inline\n",
+    "\n",
+    "from matplotlib.pyplot import imshow\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "from scipy import misc\n",
+    "\n",
+    "import os\n",
+    "import random\n",
+    "import pickle"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['0', '1']\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "imageFolder = \"-catsdogs\"\n",
+    "\n",
+    "folders = os.listdir(imageFolder)\n",
+    "num_categories = len(folders)\n",
+    "\n",
+    "print folders"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Load data complete\n"
+     ]
+    }
+   ],
+   "source": [
+    "# specify desired image properties\n",
+    "# in this case we want black and white square images 64x64 pixels in size\n",
+    "image_dim = 1 # black and white\n",
+    "image_size = 64\n",
+    "\n",
+    "# create an empty array to store the image data\n",
+    "data = []\n",
+    "\n",
+    "# look inside each folder which represents the categories of our data\n",
+    "for folder in folders:\n",
+    "    \n",
+    "    # find the files within each folder\n",
+    "    fileNames = os.listdir(\"/\".join([imageFolder, folder]))\n",
+    "    \n",
+    "    # for each file, load and process each image\n",
+    "    # in this case we limit the number of images used per cateogry to 10,000\n",
+    "    # to prevent overloading our RAM memory\n",
+    "    for fileName in fileNames[:10000]:\n",
+    "        \n",
+    "        # read in the image data into a numpy array\n",
+    "        img = misc.imread(\"/\".join([imageFolder, folder, fileName]))\n",
+    "        \n",
+    "        # if the image contains more than one color channel,\n",
+    "        # take only the first channel (in effect, convert it to black and white)\n",
+    "        if image_dim == 1 and len(img.shape) > 2: \n",
+    "            img = img[:,:,0] # convert to black and white\n",
+    "\n",
+    "        # resize to target resolution if necessary\n",
+    "        if img.shape[0] != image_size or img.shape[1] != image_size:\n",
+    "            img = misc.imresize(img, (image_size, image_size), interp='nearest')\n",
+    "\n",
+    "        # normalize data to have mean 0 and standard deviation 1\n",
+    "        # then rescale it to roughly the range 0-1\n",
+    "        img = (img - img.mean()) / img.std() / 4 + 0.5\n",
+    "        \n",
+    "        # add the image data and the associated category \n",
+    "        # (which is stored in the folder variable) to the data set\n",
+    "        # for this to work you need to make sure your folders \n",
+    "        # are named sequentially starting with 0\n",
+    "        data.append([img, folder])\n",
+    "\n",
+    "print \"Load data complete\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "\n",
+    "random.shuffle(data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "X = np.ndarray((len(data), image_size, image_size), dtype=np.float32)\n",
+    "y = np.ndarray((len(data), 1), dtype=np.int32)\n",
+    "\n",
+    "for i, d in enumerate(data):\n",
+    "    X[i] = d[0]\n",
+    "    y[i] = d[1]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "image dimensions: (64, 64)\n",
+      "target category: cat\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFiCAYAAAAna2l5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niE5YRKQjOmERkY7ohEVEOvK/bREV0im4+o4AAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7ff846af1c90>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "img_index = 2\n",
+    "\n",
+    "img = X[img_index]\n",
+    "print \"image dimensions:\", img.shape\n",
+    "print \"target category:\", (['cat', 'dog'][y[img_index][0]])\n",
+    "\n",
+    "imshow(img, cmap = plt.get_cmap('gray'), vmin = 0, vmax = 1,  interpolation='nearest')\n",
+    "plt.axis('off')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "\n",
+    "trainingSplit = int(.7 * X.shape[0])\n",
+    "\n",
+    "X_train = X[:trainingSplit]\n",
+    "y_train = y[:trainingSplit]\n",
+    "X_test = X[trainingSplit:]\n",
+    "y_test = y[trainingSplit:]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved data to -catsdogs.pickle\n",
+      "Compressed pickle size: 327760375\n"
+     ]
+    }
+   ],
+   "source": [
+    "pickle_file = imageFolder + '.pickle'\n",
+    "\n",
+    "try:\n",
+    "    f = open(pickle_file, 'wb')\n",
+    "    save = {\n",
+    "        'X_train': X_train,\n",
+    "        'y_train': y_train,\n",
+    "        'X_test': X_test,\n",
+    "        'y_test': y_test,\n",
+    "    }\n",
+    "    pickle.dump(save, f, pickle.HIGHEST_PROTOCOL)\n",
+    "    f.close()\n",
+    "except Exception as e:\n",
+    "    print 'Unable to save data to', pickle_file, ':', e\n",
+    "    raise\n",
+    "    \n",
+    "statinfo = os.stat(pickle_file)\n",
+    "print 'Saved data to', pickle_file\n",
+    "print 'Compressed pickle size:', statinfo.st_size"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-5/lab  5.1.ipynb b/notebooks/week-5/lab  5.1.ipynb
new file mode 100644
index 0000000..99abab7
--- /dev/null
+++ b/notebooks/week-5/lab  5.1.ipynb	
@@ -0,0 +1,413 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "using ordering: tf\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "import numpy as np\n",
+    "np.random.seed(1337)  # for reproducibility\n",
+    "\n",
+    "from keras.datasets import mnist\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense, Dropout, Activation, Flatten\n",
+    "from keras.layers import Convolution2D, MaxPooling2D\n",
+    "from keras.utils import np_utils\n",
+    "from keras import backend as K\n",
+    "\n",
+    "from keras.datasets import mnist\n",
+    "\n",
+    "import pickle\n",
+    "\n",
+    "print \"using ordering:\", K.image_dim_ordering()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.pkl.gz\n",
+      "15294464/15296311 [============================>.] - ETA: 0sX_train shape: (60000, 28, 28)\n",
+      "60000 train samples\n",
+      "10000 test samples\n"
+     ]
+    }
+   ],
+   "source": [
+    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
+    "\n",
+    "print 'X_train shape:', X_train.shape\n",
+    "print X_train.shape[0], 'train samples'\n",
+    "print X_test.shape[0], 'test samples'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "X_train = X_train.astype('float32')\n",
+    "X_test = X_test.astype('float32')\n",
+    "X_train /= 255.0\n",
+    "X_test /= 255.0"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(60000, 28, 28, 1)\n",
+      "(60000,)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# number of classes\n",
+    "num_classes = 10\n",
+    "\n",
+    "# image dimensions\n",
+    "img_rows, img_cols = X_train.shape[1],  X_train.shape[2]\n",
+    "\n",
+    "if K.image_dim_ordering() == 'th':\n",
+    "    X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)\n",
+    "    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)\n",
+    "    input_shape = (1, img_rows, img_cols)\n",
+    "else:\n",
+    "    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
+    "    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
+    "    input_shape = (img_rows, img_cols, 1)\n",
+    "\n",
+    "Y_train = np_utils.to_categorical(y_train, num_classes)\n",
+    "Y_test = np_utils.to_categorical(y_test, num_classes)\n",
+    "\n",
+    "print X_train.shape\n",
+    "print y_train.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(28, 28)\n",
+      "(60000,)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7f3a0150ed50>"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFfCAYAAACfj30KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "text/plain": [
+       "<matplotlib.figure.Figure at 0x7f3a116fe090>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "from matplotlib.pyplot import imshow\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "img_num = 0\n",
+    "\n",
+    "if K.image_dim_ordering() == 'th':\n",
+    "    img = X_train[img_num][0,:,:]\n",
+    "else:\n",
+    "    img = X_train[img_num][:,:,0]\n",
+    "\n",
+    "print img.shape\n",
+    "print y_train.shape\n",
+    "imshow(img, cmap = plt.get_cmap('gray'), vmin = 0, vmax = 1,  interpolation='nearest')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# model hyperparameters\n",
+    "batch_size = 128\n",
+    "nb_epoch = 30\n",
+    "\n",
+    "# network architecture\n",
+    "patch_size_1 = 5\n",
+    "patch_size_2 = 5\n",
+    "\n",
+    "depth_1 = 20\n",
+    "depth_2 = 40\n",
+    "\n",
+    "pool_size = 2\n",
+    "\n",
+    "num_hidden_1 = 1000\n",
+    "num_hidden_2 = 1000\n",
+    "\n",
+    "dropout = 0.5"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# create new Keras Sequential model\n",
+    "model = Sequential()\n",
+    "\n",
+    "# add first convolutional layer to model and specify it's depth and filter size\n",
+    "# for the first layer we also have to specify the size of each input image\n",
+    "# which we calculated above\n",
+    "model.add(Convolution2D(depth_1, patch_size_1, patch_size_1,\n",
+    "                        border_mode='valid',\n",
+    "                        input_shape=input_shape))\n",
+    "# apply 'relu' activation function for first layer\n",
+    "model.add(Activation('relu'))\n",
+    "# apply max pooling to reduce the size of the image by a factor of 2\n",
+    "model.add(MaxPooling2D(pool_size=(pool_size, pool_size)))\n",
+    "\n",
+    "# repeat these operations for the second convolutional layer\n",
+    "# this time Keras can figure out the input size \n",
+    "# from the previous layer on it's own\n",
+    "model.add(Convolution2D(depth_2, patch_size_2, patch_size_2,\n",
+    "                        border_mode='valid'))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(MaxPooling2D(pool_size=(pool_size, pool_size)))\n",
+    "\n",
+    "# flatten the three-dimensional convolutional layer to a single layer of neurons\n",
+    "model.add(Flatten())\n",
+    "\n",
+    "# add the first fully connected layer, applying 'relu' activation and dropout\n",
+    "model.add(Dense(num_hidden_1))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(Dropout(dropout))\n",
+    "\n",
+    "# add the second fully connected layer\n",
+    "model.add(Dense(num_hidden_2))\n",
+    "model.add(Activation('relu'))\n",
+    "model.add(Dropout(dropout))\n",
+    "\n",
+    "# add the final classification layer with the number of neurons \n",
+    "# matching the number of classes we are trying to learn\n",
+    "model.add(Dense(num_classes))\n",
+    "\n",
+    "# apply the 'softmax' activation to the final layer to convert the output to \n",
+    "# a probability distribution\n",
+    "model.add(Activation('softmax'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "model.compile(loss='categorical_crossentropy',\n",
+    "              optimizer='adadelta',\n",
+    "              metrics=['accuracy'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Train on 60000 samples, validate on 10000 samples\n",
+      "Epoch 1/30\n",
+      "60000/60000 [==============================] - 94s - loss: 0.2919 - acc: 0.9079 - val_loss: 0.0641 - val_acc: 0.9797\n",
+      "Epoch 2/30\n",
+      "60000/60000 [==============================] - 90s - loss: 0.0794 - acc: 0.9755 - val_loss: 0.0401 - val_acc: 0.9859\n",
+      "Epoch 3/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0567 - acc: 0.9827 - val_loss: 0.0310 - val_acc: 0.9891\n",
+      "Epoch 4/30\n",
+      "60000/60000 [==============================] - 86s - loss: 0.0459 - acc: 0.9856 - val_loss: 0.0262 - val_acc: 0.9908\n",
+      "Epoch 5/30\n",
+      "60000/60000 [==============================] - 85s - loss: 0.0374 - acc: 0.9884 - val_loss: 0.0246 - val_acc: 0.9914\n",
+      "Epoch 6/30\n",
+      "60000/60000 [==============================] - 86s - loss: 0.0314 - acc: 0.9902 - val_loss: 0.0246 - val_acc: 0.9919\n",
+      "Epoch 7/30\n",
+      "60000/60000 [==============================] - 85s - loss: 0.0276 - acc: 0.9913 - val_loss: 0.0271 - val_acc: 0.9913\n",
+      "Epoch 8/30\n",
+      "60000/60000 [==============================] - 86s - loss: 0.0246 - acc: 0.9922 - val_loss: 0.0229 - val_acc: 0.9929\n",
+      "Epoch 9/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0209 - acc: 0.9933 - val_loss: 0.0225 - val_acc: 0.9923\n",
+      "Epoch 10/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0190 - acc: 0.9939 - val_loss: 0.0206 - val_acc: 0.9926\n",
+      "Epoch 11/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0172 - acc: 0.9945 - val_loss: 0.0221 - val_acc: 0.9935\n",
+      "Epoch 12/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0160 - acc: 0.9948 - val_loss: 0.0269 - val_acc: 0.9920\n",
+      "Epoch 13/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0136 - acc: 0.9956 - val_loss: 0.0212 - val_acc: 0.9932\n",
+      "Epoch 14/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0132 - acc: 0.9956 - val_loss: 0.0214 - val_acc: 0.9932\n",
+      "Epoch 15/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0117 - acc: 0.9966 - val_loss: 0.0204 - val_acc: 0.9935\n",
+      "Epoch 16/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0106 - acc: 0.9968 - val_loss: 0.0223 - val_acc: 0.9929\n",
+      "Epoch 17/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0095 - acc: 0.9973 - val_loss: 0.0211 - val_acc: 0.9937\n",
+      "Epoch 18/30\n",
+      "60000/60000 [==============================] - 87s - loss: 0.0088 - acc: 0.9974 - val_loss: 0.0224 - val_acc: 0.9937\n",
+      "Epoch 19/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0083 - acc: 0.9974 - val_loss: 0.0237 - val_acc: 0.9939\n",
+      "Epoch 20/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0076 - acc: 0.9977 - val_loss: 0.0220 - val_acc: 0.9941\n",
+      "Epoch 21/30\n",
+      "60000/60000 [==============================] - 89s - loss: 0.0072 - acc: 0.9977 - val_loss: 0.0257 - val_acc: 0.9921\n",
+      "Epoch 22/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0065 - acc: 0.9980 - val_loss: 0.0235 - val_acc: 0.9941\n",
+      "Epoch 23/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0063 - acc: 0.9982 - val_loss: 0.0231 - val_acc: 0.9942\n",
+      "Epoch 24/30\n",
+      "60000/60000 [==============================] - 89s - loss: 0.0054 - acc: 0.9983 - val_loss: 0.0232 - val_acc: 0.9942\n",
+      "Epoch 25/30\n",
+      "60000/60000 [==============================] - 89s - loss: 0.0054 - acc: 0.9982 - val_loss: 0.0247 - val_acc: 0.9933\n",
+      "Epoch 26/30\n",
+      "60000/60000 [==============================] - 99s - loss: 0.0044 - acc: 0.9988 - val_loss: 0.0298 - val_acc: 0.9926\n",
+      "Epoch 27/30\n",
+      "60000/60000 [==============================] - 125s - loss: 0.0045 - acc: 0.9987 - val_loss: 0.0239 - val_acc: 0.9933\n",
+      "Epoch 28/30\n",
+      "60000/60000 [==============================] - 197s - loss: 0.0042 - acc: 0.9987 - val_loss: 0.0266 - val_acc: 0.9928\n",
+      "Epoch 29/30\n",
+      "60000/60000 [==============================] - 91s - loss: 0.0036 - acc: 0.9990 - val_loss: 0.0249 - val_acc: 0.9937\n",
+      "Epoch 30/30\n",
+      "60000/60000 [==============================] - 88s - loss: 0.0036 - acc: 0.9989 - val_loss: 0.0236 - val_acc: 0.9941\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<keras.callbacks.History at 0x7f39f8422690>"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.fit(X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch,\n",
+    "          verbose=1, validation_data=(X_test, Y_test))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test score: 0.0236249855404\n",
+      "Test accuracy: 99.41%\n"
+     ]
+    }
+   ],
+   "source": [
+    "score = model.evaluate(X_test, Y_test, verbose=0)\n",
+    "\n",
+    "print 'Test score:', score[0]\n",
+    "print 'Test accuracy: {:.2%}'.format(score[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-6/6.2.ipynb b/notebooks/week-6/6.2.ipynb
new file mode 100644
index 0000000..ee5ee88
--- /dev/null
+++ b/notebooks/week-6/6.2.ipynb
@@ -0,0 +1,186 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense\n",
+    "from keras.layers import Dropout\n",
+    "from keras.layers import LSTM\n",
+    "from keras.callbacks import ModelCheckpoint\n",
+    "from keras.utils import np_utils\n",
+    "\n",
+    "import sys\n",
+    "import re\n",
+    "import pickle"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Training set', (18212, 100, 44), (18212, 44))\n"
+     ]
+    }
+   ],
+   "source": [
+    "pickle_file = '-basic_data.pickle'\n",
+    "\n",
+    "with open(pickle_file, 'rb') as f:\n",
+    "    save = pickle.load(f)\n",
+    "    X = save['X']\n",
+    "    y = save['y']\n",
+    "    char_to_int = save['char_to_int']  \n",
+    "    int_to_char = save['int_to_char']    \n",
+    "    del save  # hint to help gc free up memory\n",
+    "    print('Training set', X.shape, y.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "\n",
+    "# define the LSTM model\n",
+    "model = Sequential()\n",
+    "model.add(LSTM(128, return_sequences=False, input_shape=(X.shape[1], X.shape[2])))\n",
+    "# model.add(Dropout(0.50))\n",
+    "model.add(Dense(y.shape[1], activation='softmax'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# load the parameters from the pretrained model\n",
+    "filename = \"-basic_LSTM.hdf5\"\n",
+    "model.load_weights(filename)\n",
+    "model.compile(loss='categorical_crossentropy', optimizer='adam')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def sample(preds, temperature=1.0):\n",
+    "    preds = np.asarray(preds).astype('float64')\n",
+    "    preds = np.log(preds) / temperature\n",
+    "    exp_preds = np.exp(preds)\n",
+    "    preds = exp_preds / np.sum(exp_preds)\n",
+    "    probas = np.random.multinomial(1, preds, 1)\n",
+    "    return np.argmax(probas)\n",
+    "def generate(sentence, sample_length=50, diversity=0.35):\n",
+    "    generated = sentence\n",
+    "    sys.stdout.write(generated)\n",
+    "\n",
+    "    for i in range(sample_length):\n",
+    "        x = np.zeros((1, X.shape[1], X.shape[2]))\n",
+    "        for t, char in enumerate(sentence):\n",
+    "            x[0, t, char_to_int[char]] = 1.\n",
+    "\n",
+    "        preds = model.predict(x, verbose=0)[0]\n",
+    "        next_index = sample(preds, diversity)\n",
+    "        next_char = int_to_char[next_index]\n",
+    "\n",
+    "        generated += next_char\n",
+    "        sentence = sentence[1:] + next_char\n",
+    "\n",
+    "        sys.stdout.write(next_char)\n",
+    "        sys.stdout.flush()\n",
+    "    print"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "america has shown that progress is possible. last year, income gains were larger for households at the lome and comingitat om now rutce to the rate tha the workers cann a poricis and best conngriss of the some that be and raves con the partical of expart for to mes ic thase sthe poritifil ny wore the in at the mant, and to ce sore the growit sever and emertecine suthen frererom the 197, thine sicaniss for the progress recound in or-ercons, growth, and and in part date real ind coring the pastorics sothe that as in as the pored to were und roster the prestermes that chares of and lising of econ\n",
+      "--------------------\n",
+      "and as people around the world began to hear the tale of the lowly colonists who overthrew an empiresto bould in of reale so pordest of the fince of the sed consent we the partican and in marins the wills and well and porter in for and resersts and a corpeas and economic songe. in the pation and nesing the wore more the wors of ale and amportote so pors wilk sing conting the wher poritical inceat in the eas and of ment dest a domingr pat tive mat cound in americans for and incecono sofrecs frow les proged the pratical canity mens rowe best of the some the fere a duticil seving in economy decon\n",
+      "--------------------\n"
+     ]
+    }
+   ],
+   "source": [
+    "prediction_length = 500\n",
+    "seed_from_text = \"america has shown that progress is possible. last year, income gains were larger for households at t\"\n",
+    "seed_original = \"and as people around the world began to hear the tale of the lowly colonists who overthrew an empire\"\n",
+    "\n",
+    "for seed in [seed_from_text, seed_original]:\n",
+    "    generate(seed, prediction_length, .50)\n",
+    "    print \"-\" * 20"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-6/assignment 6.1.ipynb b/notebooks/week-6/assignment 6.1.ipynb
new file mode 100644
index 0000000..4a9bc42
--- /dev/null
+++ b/notebooks/week-6/assignment 6.1.ipynb	
@@ -0,0 +1,964 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense\n",
+    "from keras.layers import Dropout\n",
+    "from keras.layers import LSTM\n",
+    "from keras.callbacks import ModelCheckpoint\n",
+    "from keras.utils import np_utils\n",
+    "\n",
+    "from time import gmtime, strftime\n",
+    "import os\n",
+    "import re\n",
+    "import pickle\n",
+    "import random\n",
+    "import sys"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "length of text: 18312\n",
+      "text preview: wherever i go these days, at home or abroad, people ask me the same question: what is happening in the american political system? how has a country that has benefitedperhaps more than any otherfrom immigration, trade and technological innovation suddenly developed a strain of anti-immigrant, anti-innovation protectionism? why have some on the far left and even more on the far right embraced a crude populism that promises a return to a past that is not possible to restoreand that, for most americ\n"
+     ]
+    }
+   ],
+   "source": [
+    "# load ascii text from file\n",
+    "filename = \"data/obama.txt\"\n",
+    "raw_text = open(filename).read()\n",
+    "\n",
+    "# get rid of any characters other than letters, numbers, \n",
+    "# and a few special characters\n",
+    "raw_text = re.sub('[^\\nA-Za-z0-9 ,.:;?!-]+', '', raw_text)\n",
+    "\n",
+    "# convert all text to lowercase\n",
+    "raw_text = raw_text.lower()\n",
+    "\n",
+    "n_chars = len(raw_text)\n",
+    "print \"length of text:\", n_chars\n",
+    "print \"text preview:\", raw_text[:500]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "number of unique characters found: 44\n",
+      "a - maps to -> 18\n",
+      "25 - maps to -> h\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "# extract all unique characters in the text\n",
+    "chars = sorted(list(set(raw_text)))\n",
+    "n_vocab = len(chars)\n",
+    "print \"number of unique characters found:\", n_vocab\n",
+    "\n",
+    "# create mapping of characters to integers and back\n",
+    "char_to_int = dict((c, i) for i, c in enumerate(chars))\n",
+    "int_to_char = dict((i, c) for i, c in enumerate(chars))\n",
+    "\n",
+    "# test our mapping\n",
+    "print 'a', \"- maps to ->\", char_to_int[\"a\"]\n",
+    "print 25, \"- maps to ->\", int_to_char[25]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Total sequences:  18212\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "# prepare the dataset of input to output pairs encoded as integers\n",
+    "seq_length = 100\n",
+    "\n",
+    "inputs = []\n",
+    "outputs = []\n",
+    "\n",
+    "for i in range(0, n_chars - seq_length, 1):\n",
+    "    inputs.append(raw_text[i:i + seq_length])\n",
+    "    outputs.append(raw_text[i + seq_length])\n",
+    "    \n",
+    "n_sequences = len(inputs)\n",
+    "print \"Total sequences: \", n_sequences"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "\n",
+    "indeces = range(len(inputs))\n",
+    "random.shuffle(indeces)\n",
+    "\n",
+    "inputs = [inputs[x] for x in indeces]\n",
+    "outputs = [outputs[x] for x in indeces]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "ore prosperous than ever before and yet our societies are marked by uncertainty and unease. so we ha --> v\n"
+     ]
+    }
+   ],
+   "source": [
+    "print inputs[0], \"-->\", outputs[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "X dims --> (18212, 100, 44)\n",
+      "y dims --> (18212, 44)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# create two empty numpy array with the proper dimensions\n",
+    "X = np.zeros((n_sequences, seq_length, n_vocab), dtype=np.bool)\n",
+    "y = np.zeros((n_sequences, n_vocab), dtype=np.bool)\n",
+    "\n",
+    "# iterate over the data and build up the X and y data sets\n",
+    "# by setting the appropriate indices to 1 in each one-hot vector\n",
+    "for i, example in enumerate(inputs):\n",
+    "    for t, char in enumerate(example):\n",
+    "        X[i, t, char_to_int[char]] = 1\n",
+    "    y[i, char_to_int[outputs[i]]] = 1\n",
+    "    \n",
+    "print 'X dims -->', X.shape\n",
+    "print 'y dims -->', y.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# define the LSTM model\n",
+    "model = Sequential()\n",
+    "model.add(LSTM(128, return_sequences=False, input_shape=(X.shape[1], X.shape[2])))\n",
+    "model.add(Dropout(0.50))\n",
+    "model.add(Dense(y.shape[1], activation='softmax'))\n",
+    "model.compile(loss='categorical_crossentropy', optimizer='adam')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def sample(preds, temperature=1.0):\n",
+    "    # helper function to sample an index from a probability array\n",
+    "    preds = np.asarray(preds).astype('float64')\n",
+    "    preds = np.log(preds) / temperature\n",
+    "    exp_preds = np.exp(preds)\n",
+    "    preds = exp_preds / np.sum(exp_preds)\n",
+    "    probas = np.random.multinomial(1, preds, 1)\n",
+    "    return np.argmax(probas)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def generate(sentence, prediction_length=50, diversity=0.35):\n",
+    "    print '----- diversity:', diversity \n",
+    "\n",
+    "    generated = sentence\n",
+    "    sys.stdout.write(generated)\n",
+    "\n",
+    "    # iterate over number of characters requested\n",
+    "    for i in range(prediction_length):\n",
+    "        \n",
+    "        # build up sequence data from current sentence\n",
+    "        x = np.zeros((1, X.shape[1], X.shape[2]))\n",
+    "        for t, char in enumerate(sentence):\n",
+    "            x[0, t, char_to_int[char]] = 1.\n",
+    "\n",
+    "        # use trained model to return probability distribution\n",
+    "        # for next character based on input sequence\n",
+    "        preds = model.predict(x, verbose=0)[0]\n",
+    "        \n",
+    "        # use sample() function to sample next character \n",
+    "        # based on probability distribution and desired diversity\n",
+    "        next_index = sample(preds, diversity)\n",
+    "        \n",
+    "        # convert integer to character\n",
+    "        next_char = int_to_char[next_index]\n",
+    "\n",
+    "        # add new character to generated text\n",
+    "        generated += next_char\n",
+    "        \n",
+    "        # delete the first character from beginning of sentance, \n",
+    "        # and add new caracter to the end. This will form the \n",
+    "        # input sequence for the next predicted character.\n",
+    "        sentence = sentence[1:] + next_char\n",
+    "\n",
+    "        # print results to screen\n",
+    "        sys.stdout.write(next_char)\n",
+    "        sys.stdout.flush()\n",
+    "    print"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "filepath=\"-basic_LSTM.hdf5\"\n",
+    "checkpoint = ModelCheckpoint(filepath, monitor='loss', verbose=0, save_best_only=True, mode='min')\n",
+    "callbacks_list = [checkpoint]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "epoch: 1 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 125s - loss: 3.2184 - val_loss: 2.9761\n",
+      "----- generating with seed: any previous administration since at least 1960.\n",
+      "\n",
+      "even these efforts fall well short. in the future,\n",
+      "----- diversity: 0.5\n",
+      "any previous administration since at least 1960.\n",
+      "\n",
+      "even these efforts fall well short. in the future,oono t eagei h    ittst ea o iads  n dd t  e  ery aa meso eot   o ao  a t  reoet ou d  t ot e  t to \n",
+      "----- diversity: 1.2\n",
+      "any previous administration since at least 1960.\n",
+      "\n",
+      "even these efforts fall well short. in the future,o tea ottrrp    h1gochn  ce crv erts0rrriclpemh\n",
+      "a meiew srtxo?sh h.iny mnoespo e  eiit heie erjtea,h\n",
+      "epoch: 2 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 123s - loss: 3.0241 - val_loss: 2.9319\n",
+      "----- generating with seed: and the financial crisis of 2008 only seemed to increase the isolation of corporations and elites, w\n",
+      "----- diversity: 0.5\n",
+      "and the financial crisis of 2008 only seemed to increase the isolation of corporations and elites, wtn  eto ict  el eut a e  osisii a sg trnosenasata e   t   ndti h roie t e    ag e rea leat si o  tei\n",
+      "----- diversity: 1.2\n",
+      "and the financial crisis of 2008 only seemed to increase the isolation of corporations and elites, wtr 9actgs otgp bty1a e srdie9oearrn?acwttl4tl le tiee aerhvhoeitej em sooolisoohks n syikyl1 3aolts \n",
+      "epoch: 3 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 132s - loss: 2.9646 - val_loss: 2.8871\n",
+      "----- generating with seed:  anti-refugee sentiment expressed by some americans today echoes nativist lurches of the pastthe ali\n",
+      "----- diversity: 0.5\n",
+      " anti-refugee sentiment expressed by some americans today echoes nativist lurches of the pastthe alie t co  e e oo f ue alms n es  ies  rod d n nos issh  ana rr  aieond  st to  ruesl e t na lh r ofhhn\n",
+      "----- diversity: 1.2\n",
+      " anti-refugee sentiment expressed by some americans today echoes nativist lurches of the pastthe ali  su hevy o8iaade wcrar,a eb inyn e i3ry uc1aaets  teb luw te ttii 1sudtsilinneryuhtsgrqeepueiosu lp\n",
+      "epoch: 4 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.8854 - val_loss: 2.7973\n",
+      "----- generating with seed: thats not only sustainable but shared. to achieve it america must stay committed to working with all\n",
+      "----- diversity: 0.5\n",
+      "thats not only sustainable but shared. to achieve it america must stay committed to working with all t ot , din as tetwriem tes oanr i o r rnas t an arta onr  oons  e o ay  ene ni son o mm are io  oo \n",
+      "----- diversity: 1.2\n",
+      "thats not only sustainable but shared. to achieve it america must stay committed to working with allmcn oysntaddr\n",
+      "is ppha oibn oldpon h s,ieddhod amtsepeinu r tomwro ce ps syuoi kald fy m iroghwtas tc\n",
+      "epoch: 5 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.7802 - val_loss: 2.6728\n",
+      "----- generating with seed:  new rules cutting emissions from vehicles and power plants.\n",
+      "\n",
+      "the results are clear: a more durable,\n",
+      "----- diversity: 0.5\n",
+      " new rules cutting emissions from vehicles and power plants.\n",
+      "\n",
+      "the results are clear: a more durable,  oe as thoe ibe tole an. oe olss pre ai s eu errtde toe toa  att sae to a tue oo roenp on ao t are \n",
+      "----- diversity: 1.2\n",
+      " new rules cutting emissions from vehicles and power plants.\n",
+      "\n",
+      "the results are clear: a more durable,pvwhtt prto 6enyioami  nfmw dcuct ianbcle re ay\n",
+      "it,tto oalgts ted \n",
+      "racboafaa deauast has0lsalregcu.i\n",
+      "epoch: 6 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.6767 - val_loss: 2.5923\n",
+      "----- generating with seed: ch, at their childrens schools, in civic organisations. thats why ceos took home about 20- to 30-tim\n",
+      "----- diversity: 0.5\n",
+      "ch, at their childrens schools, in civic organisations. thats why ceos took home about 20- to 30-tims ar toal enfures coolins othant onthen eare ifue\n",
+      "anr aca ae norens on ore es torase and erae pherre\n",
+      "----- diversity: 1.2\n",
+      "ch, at their childrens schools, in civic organisations. thats why ceos took home about 20- to 30-timftareleny.. hagisesonupptos anmfrold nthesscpaaa imoind ethe- nowthet enr racroisdeonnma\n",
+      "af227rfphcn\n",
+      "epoch: 7 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 114s - loss: 2.5954 - val_loss: 2.5025\n",
+      "----- generating with seed: nsion than many appreciated in recovering from our crisismore than a dozen bills provided 1.4 trilli\n",
+      "----- diversity: 0.5\n",
+      "nsion than many appreciated in recovering from our crisismore than a dozen bills provided 1.4 trillin more ores or ire an ee the the ntot the as an re and an  antere mont the and se tho as ron the the\n",
+      "----- diversity: 1.2\n",
+      "nsion than many appreciated in recovering from our crisismore than a dozen bills provided 1.4 trillinerlsvioite ricdionwt innrove\n",
+      " neigatttnochywvi\n",
+      "y. ue anh ones,wadd nhle lurung feysznag  ppowtee. v\n",
+      "epoch: 8 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 116s - loss: 2.5312 - val_loss: 2.4436\n",
+      "----- generating with seed:  precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on h\n",
+      "----- diversity: 0.5\n",
+      " precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on her porle lha tor te the rerise or cell oar  oor iaitd ao che tho aa the the apntol sin sorte poreste\n",
+      "----- diversity: 1.2\n",
+      " precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on hst arolss frong crovi1fa nl hals hhcoual, uo ,wehumte fovs g bdorpsatweme\n",
+      "is so. rydven i s ho iamui\n",
+      "epoch: 9 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 2.4893 - val_loss: 2.4156\n",
+      "----- generating with seed:  with health insurance, while health-care costs grow at the slowest rate in 50 years; annual deficit\n",
+      "----- diversity: 0.5\n",
+      " with health insurance, while health-care costs grow at the slowest rate in 50 years; annual deficitit ans iot berion in the rorames the pant monture the ante. rofore s shesincmed no tererat 1oro tha \n",
+      "----- diversity: 1.2\n",
+      " with health insurance, while health-care costs grow at the slowest rate in 50 years; annual deficite  om\n",
+      "fgust tha1d wuirmerbilemth oa; wolecse,tanif h armed,eiottoaga bendd, otem to cgony fnliclxluc\n",
+      "epoch: 10 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.4497 - val_loss: 2.3785\n",
+      "----- generating with seed: doing it. and those who should be rising in defence of further reform too often ignore the progress \n",
+      "----- diversity: 0.5\n",
+      "doing it. and those who should be rising in defence of further reform too often ignore the progress in core the pand the nomite the to the roalis ro smering on icat ipgeres on urete meftore s and meri\n",
+      "----- diversity: 1.2\n",
+      "doing it. and those who should be rising in defence of further reform too often ignore the progress ,s mite cer fiyt oin fo inscentlians os w\n",
+      "yjy fae. hecond tooww yddqpahe sgiomd noedutoftriang thele\n",
+      "epoch: 11 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 2.4139 - val_loss: 2.3523\n",
+      "----- generating with seed: em since the 1930s, as well as reforming health care and introducing new rules cutting emissions fro\n",
+      "----- diversity: 0.5\n",
+      "em since the 1930s, as well as reforming health care and introducing new rules cutting emissions fronis ins uracinge the faris re weris chale sithe of r af forly an the word te mang mare an se chuto r\n",
+      "----- diversity: 1.2\n",
+      "em since the 1930s, as well as reforming health care and introducing new rules cutting emissions frole eewan staneon thap honvereupsxnora 2rtas meromrensu ect phacing lfr uvulg. ighe ienthprlendmig um\n",
+      "epoch: 12 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 116s - loss: 2.3855 - val_loss: 2.3226\n",
+      "----- generating with seed: rsal in inequality; 20m more americans with health insurance, while health-care costs grow at the sl\n",
+      "----- diversity: 0.5\n",
+      "rsal in inequality; 20m more americans with health insurance, while health-care costs grow at the slating ond yat an then wores and the the pheobors who le inle oo the an the thal dealing or orervere \n",
+      "----- diversity: 1.2\n",
+      "rsal in inequality; 20m more americans with health insurance, while health-care costs grow at the slpoolt, eor uevatrel uomth az prultortt aprin8 tharmasses fosisc2by. cwshelsue8dhlc7 cavey nud. of ty\n",
+      "epoch: 13 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 2.3563 - val_loss: 2.2988\n",
+      "----- generating with seed: ity to secure a decent wage. too many potential physicists and engineers spend their careers shiftin\n",
+      "----- diversity: 0.5\n",
+      "ity to secure a decent wage. too many potential physicists and engineers spend their careers shifting tes an the mere they the sing or ame res the  hore be mers wer sace fore pore cons gnob ind ment a\n",
+      "----- diversity: 1.2\n",
+      "ity to secure a decent wage. too many potential physicists and engineers spend their careers shiftin tipand ,n pqeredan p-esed forle recha enum unlist, fren ty rimaco tofporkcgaor fcencu;twefn tr t4re\n",
+      "epoch: 14 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.3337 - val_loss: 2.2698\n",
+      "----- generating with seed: c dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible technological advances through the inter\n",
+      "----- diversity: 0.5\n",
+      "c dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible technological advances through the intere boticing the the tin baald the int al inger the seres lo the fars an in the the the the the and al\n",
+      "----- diversity: 1.2\n",
+      "c dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible technological advances through the interuntesing nohea stine.au oqt nese. mruceanbeg, eitace f hprat weletren cariing nr beriecre nner let c\n",
+      "epoch: 15 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 115s - loss: 2.3136 - val_loss: 2.2540\n",
+      "----- generating with seed: can get one and building a resilient economy thats primed for future growth.\n",
+      "\n",
+      "restoring economic dyn\n",
+      "----- diversity: 0.5\n",
+      "can get one and building a resilient economy thats primed for future growth.\n",
+      "\n",
+      "restoring economic dyn the prote the the the ereast on priliting al the suent the the rerole sorses om tor am the oncent o\n",
+      "----- diversity: 1.2\n",
+      "can get one and building a resilient economy thats primed for future growth.\n",
+      "\n",
+      "restoring economic dynens th aherercet 7pthet wars wactdes,d,e sorkemst tingwf, isyrmans pas coulgiemesl of rcagn cy bnamg\n",
+      "epoch: 16 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 124s - loss: 2.2891 - val_loss: 2.2401\n",
+      "----- generating with seed: alculations by the department of the treasury. while the top 1 of households now pay more of their f\n",
+      "----- diversity: 0.5\n",
+      "alculations by the department of the treasury. while the top 1 of households now pay more of their fore the soning iat en the ing oft our ness to the suris in the the far angertad and tie thut cong th\n",
+      "----- diversity: 1.2\n",
+      "alculations by the department of the treasury. while the top 1 of households now pay more of their fort bly.\n",
+      " ood blt b pirgkovimebcol butooty emerlishwh axsinc oketat donc \n",
+      "ecoconte tred.\n",
+      " thkit s,nt\n",
+      "epoch: 17 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 162s - loss: 2.2695 - val_loss: 2.2212\n",
+      "----- generating with seed:  growth and for ensuring that it is shared broadly. these include everything from boosting funding f\n",
+      "----- diversity: 0.5\n",
+      " growth and for ensuring that it is shared broadly. these include everything from boosting funding fard and ant ingreasing it om the lons pirtereis on wect on the the and betinge the same and the ald \n",
+      "----- diversity: 1.2\n",
+      " growth and for ensuring that it is shared broadly. these include everything from boosting funding for ercin yeruy sofinbure singeald rosatto.s\n",
+      "astadsube itowly iter,asit se9seufgrion ams asersctofiga\n",
+      "epoch: 18 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 208s - loss: 2.2532 - val_loss: 2.2086\n",
+      "----- generating with seed: andable frustration, much of it fanned by politicians who would actually make the problem worse rath\n",
+      "----- diversity: 0.5\n",
+      "andable frustration, much of it fanned by politicians who would actually make the problem worse rath en thar and the the an reate the tor eraling the how wing bete surmich pratist and reden cond of th\n",
+      "----- diversity: 1.2\n",
+      "andable frustration, much of it fanned by politicians who would actually make the problem worse rath idite, tun sucat jhyw re2ur tho thasvircaco., mhegt racoistalran\n",
+      "iund p thenimianl ofnogrial. oit r\n",
+      "epoch: 19 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 155s - loss: 2.2226 - val_loss: 2.1899\n",
+      "----- generating with seed: here does my successor go from here?\n",
+      "\n",
+      "further progress requires recognising that americas economy is\n",
+      "----- diversity: 0.5\n",
+      "here does my successor go from here?\n",
+      "\n",
+      "further progress requires recognising that americas economy is wee recont mer the the of recaning the porlemasting and the pared the tho the pore cor imeriand the\n",
+      "----- diversity: 1.2\n",
+      "here does my successor go from here?\n",
+      "\n",
+      "further progress requires recognising that americas economy is..lion all iparuntilysi g1kut ipheistyre, hyncr-blionw 08em nerto me sefcametiall ens om icowttymisv\n",
+      "epoch: 20 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 159s - loss: 2.2115 - val_loss: 2.1725\n",
+      "----- generating with seed: s of the shadow banking system still present vulnerabilities and the housing-finance system has not \n",
+      "----- diversity: 0.5\n",
+      "s of the shadow banking system still present vulnerabilities and the housing-finance system has not pore coute ne rati af ing ang out ono ficos and the the to an on pooll to wor wort and ous ancing th\n",
+      "----- diversity: 1.2\n",
+      "s of the shadow banking system still present vulnerabilities and the housing-finance system has not youblaty ypaltondicinoicaeum acadrakend ledereseluivesin cmanot forecend wixrulitadringire a douldsi\n",
+      "epoch: 21 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 149s - loss: 2.1929 - val_loss: 2.1697\n",
+      "----- generating with seed: d a falling minimum wage. there is something to all of these and weve made real progress on all thes\n",
+      "----- diversity: 0.5\n",
+      "d a falling minimum wage. there is something to all of these and weve made real progress on all thes the parestere in or andericingrit on wer ther core on the fores in anmertsen chest amer theis of th\n",
+      "----- diversity: 1.2\n",
+      "d a falling minimum wage. there is something to all of these and weve made real progress on all thes tecpules-at amedemon layivpipsgomtst g93 ty tap  rouceidad icreeshinged aty,launs rove thoalithin a\n",
+      "epoch: 22 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 186s - loss: 2.1703 - val_loss: 2.1464\n",
+      "----- generating with seed:  projected to earn over 8m per yearthe top 0.1by nearly 7 percentage points, based on calculations b\n",
+      "----- diversity: 0.5\n",
+      " projected to earn over 8m per yearthe top 0.1by nearly 7 percentage points, based on calculations bes the partice and pating coure and rese sutte the and anterases th abe one sante in to in the eonde\n",
+      "----- diversity: 1.2\n",
+      " projected to earn over 8m per yearthe top 0.1by nearly 7 percentage points, based on calculations bes aver fhect mpoftinsi-ta, aase ads ale thent o?th arst-evon, al uhay .keve askering-fthe song 15k \n",
+      "epoch: 23 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 181s - loss: 2.1547 - val_loss: 2.1389\n",
+      "----- generating with seed: med for future growth.\n",
+      "\n",
+      "restoring economic dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible\n",
+      "----- diversity: 0.5\n",
+      "med for future growth.\n",
+      "\n",
+      "restoring economic dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible desunt hand and the pored and por of of the sure the the resingr that peranom an sowe be for of the\n",
+      "----- diversity: 1.2\n",
+      "med for future growth.\n",
+      "\n",
+      "restoring economic dynamism\n",
+      "\n",
+      "first, in recent years, we have seen incredible loyise ahall bose of phopceg.\n",
+      "aleincibe, iny cove conatadl eyd wthuviric sorle bt alm chersthat dha\n",
+      "epoch: 24 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 195s - loss: 2.1341 - val_loss: 2.1260\n",
+      "----- generating with seed: action, self-esteem, physical health and mortality. it is related to a devastating rise of opioid ab\n",
+      "----- diversity: 0.5\n",
+      "action, self-esteem, physical health and mortality. it is related to a devastating rise of opioid ablting thas to mere the the prowing the aress mort picling bates and the of the pronte con the poromi\n",
+      "----- diversity: 1.2\n",
+      "action, self-esteem, physical health and mortality. it is related to a devastating rise of opioid abitegt poldmals allx ini.gene movuy byininad fowilh wort poribey: ap bhtpuost,ol\n",
+      "mesinos wh th suetma\n",
+      "epoch: 25 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 177s - loss: 2.1197 - val_loss: 2.1185\n",
+      "----- generating with seed: bridge and airport upgrades are nonstarters.\n",
+      "\n",
+      "we could also help private investment and innovation w\n",
+      "----- diversity: 0.5\n",
+      "bridge and airport upgrades are nonstarters.\n",
+      "\n",
+      "we could also help private investment and innovation wat the and andice tis the dofs the procticins and in prome sow hois rede in the a the arising that t\n",
+      "----- diversity: 1.2\n",
+      "bridge and airport upgrades are nonstarters.\n",
+      "\n",
+      "we could also help private investment and innovation winr predocoty icupro7 byoned, adonejit in martidy iturita pasiti sfursto: a invante ap owbry, y eles\n",
+      "epoch: 26 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 195s - loss: 2.1036 - val_loss: 2.1064\n",
+      "----- generating with seed: est man an equal chance to get rich with everybody else. thats the problem with increased inequality\n",
+      "----- diversity: 0.5\n",
+      "est man an equal chance to get rich with everybody else. thats the problem with increased inequality and toun were and the ald and and resing the part the proverst and end conthe be mory the tor as ra\n",
+      "----- diversity: 1.2\n",
+      "est man an equal chance to get rich with everybody else. thats the problem with increased inequality pant dont contistass d sfasd; ressernd gnog\n",
+      "hass ncmoring crrejofle ag arstinx ono jogerfal force t\n",
+      "epoch: 27 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 193s - loss: 2.0817 - val_loss: 2.1053\n",
+      "----- generating with seed:  already done, not undoing it. and those who should be rising in defence of further reform too often\n",
+      "----- diversity: 0.5\n",
+      " already done, not undoing it. and those who should be rising in defence of further reform too oftent and botee the and the semeres of the emore co wore aversabe and farest on the ofrerico fare the ad\n",
+      "----- diversity: 1.2\n",
+      " already done, not undoing it. and those who should be rising in defence of further reform too oftene wopr weicianl deriann citauleras, inseoldstitw, thasttolsaan moaldm insewe nnow ewkurt heap theh c\n",
+      "epoch: 28 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 205s - loss: 2.0617 - val_loss: 2.0888\n",
+      "----- generating with seed: logy that rejects virtually all sources of new public funding; a fixation on deficits at the expense\n",
+      "----- diversity: 0.5\n",
+      "logy that rejects virtually all sources of new public funding; a fixation on deficits at the expense fo the tho sowe deconom the for out in icanity, the wort des mere the an the and to the proversit e\n",
+      "----- diversity: 1.2\n",
+      "logy that rejects virtually all sources of new public funding; a fixation on deficits at the expenseds, w0o7 we rmeo tha thiod 5crssapletn. surd of omintitr5by espreged tha an oedonomtl theliy\n",
+      "\n",
+      "fanpel\n",
+      "epoch: 29 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 194s - loss: 2.0457 - val_loss: 2.0830\n",
+      "----- generating with seed: ress, segments of the shadow banking system still present vulnerabilities and the housing-finance sy\n",
+      "----- diversity: 0.5\n",
+      "ress, segments of the shadow banking system still present vulnerabilities and the housing-finance sy the workens the constiom and american sithe instalising the that on amipititilisis. farty that ore \n",
+      "----- diversity: 1.2\n",
+      "ress, segments of the shadow banking system still present vulnerabilities and the housing-finance sycon mepliranicy whold gendisgaty \n",
+      "fincn gomatictte o2t eroute gfontand vethengictim? coinprer evaini\n",
+      "epoch: 30 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 166s - loss: 2.0310 - val_loss: 2.0758\n",
+      "----- generating with seed: on or elimination of this constraining factor is one reason why todays ceo is now paid over 250-time\n",
+      "----- diversity: 0.5\n",
+      "on or elimination of this constraining factor is one reason why todays ceo is now paid over 250-time somer and cot bat the and be of the porteme the nome the on word reninge that dase for cons to 19e \n",
+      "----- diversity: 1.2\n",
+      "on or elimination of this constraining factor is one reason why todays ceo is now paid over 250-timeng cotcons somit charpess fios ycans manisg atby crtse pucs tost anion. conte pewd lisg ouln add as-\n",
+      "epoch: 31 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 161s - loss: 2.0206 - val_loss: 2.0659\n",
+      "----- generating with seed: ubstantially boosted measured productivity growth. over the past decade, america has enjoyed the fas\n",
+      "----- diversity: 0.5\n",
+      "ubstantially boosted measured productivity growth. over the past decade, america has enjoyed the fast al the tor con mation sonce to reand incont to part contend tict and in sumity the the for the ald\n",
+      "----- diversity: 1.2\n",
+      "ubstantially boosted measured productivity growth. over the past decade, america has enjoyed the fasc os og as bapiing\n",
+      "rive of tferb is of tedy avuthe chartt, thove wujd sipdabitimilamily 3t ptkdeocs \n",
+      "epoch: 32 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 202s - loss: 2.0085 - val_loss: 2.0564\n",
+      "----- generating with seed: controls as much wealth as the other 99 will never be stable. gaps between rich and poor are not new\n",
+      "----- diversity: 0.5\n",
+      "controls as much wealth as the other 99 will never be stable. gaps between rich and poor are not new hes more as the prostes. that to eas patily bes th the to par dy of tho cons mere ald repard of the\n",
+      "----- diversity: 1.2\n",
+      "controls as much wealth as the other 99 will never be stable. gaps between rich and poor are not new th hab ty -bd uctear oftpmanle for anterabkins and jove winhy oub elcabitites manimu, sroweltae-e s\n",
+      "epoch: 33 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 153s - loss: 1.9866 - val_loss: 2.0573\n",
+      "----- generating with seed: ity growth, combating rising inequality, ensuring that everyone who wants a job can get one and buil\n",
+      "----- diversity: 0.5\n",
+      "ity growth, combating rising inequality, ensuring that everyone who wants a job can get one and builing the with and and ihat of to mering the premerte fur inengatest fis reco encenomy that in contree\n",
+      "----- diversity: 1.2\n",
+      "ity growth, combating rising inequality, ensuring that everyone who wants a job can get one and builktung. eswalisys conorsuploted not any iteaca ve ouskarpjof enmilian in cy a thej streot of. etsema \n",
+      "epoch: 34 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 1.9643 - val_loss: 2.0511\n",
+      "----- generating with seed: tself, has taken hold in america. its not new, nor is it dissimilar to a discontent spreading throug\n",
+      "----- diversity: 0.5\n",
+      "tself, has taken hold in america. its not new, nor is it dissimilar to a discontent spreading throughit s and re ard peolly and the lofdere the wor ingreg-thoud the rester ne ind the aide ap the wort \n",
+      "----- diversity: 1.2\n",
+      "tself, has taken hold in america. its not new, nor is it dissimilar to a discontent spreading throughol ectnonst of 2hons-ticcy, cofrevar ad flahe wo thet thes, axn jome but. and auler amenntinid cis \n",
+      "epoch: 35 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 116s - loss: 1.9568 - val_loss: 2.0510\n",
+      "----- generating with seed: eties are marked by uncertainty and unease. so we have a choiceretreat into old, closed-off economie\n",
+      "----- diversity: 0.5\n",
+      "eties are marked by uncertainty and unease. so we have a choiceretreat into old, closed-off economie should and andiling-ton the more fercans rowth than econome siccnstoo inaly compresting the erores \n",
+      "----- diversity: 1.2\n",
+      "eties are marked by uncertainty and unease. so we have a choiceretreat into old, closed-off economie ounceet. cevtingriof thes and oof reoming hwaldnting cttente vinch, ios, thes latp nat. tid, linde \n",
+      "epoch: 36 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 119s - loss: 1.9342 - val_loss: 2.0522\n",
+      "----- generating with seed: f stabilising our economy. unfortunately, good economics can be overridden by bad politics. my admin\n",
+      "----- diversity: 0.5\n",
+      "f stabilising our economy. unfortunately, good economics can be overridden by bad politics. my adminingre the are conoming in the arse and reversen she pasinit and canting bution mant and toran by eas\n",
+      "----- diversity: 1.2\n",
+      "f stabilising our economy. unfortunately, good economics can be overridden by bad politics. my adminasjot enliinco a touns in nomine.\n",
+      "\n",
+      "afirg hot the ians intreant.\n",
+      "\n",
+      "aecr ails. and 2050a ssot., than we\n",
+      "epoch: 37 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 1.9201 - val_loss: 2.0347\n",
+      "----- generating with seed:  keep more americans in the labour market when they fall on hard times. these include providing wage\n",
+      "----- diversity: 0.5\n",
+      " keep more americans in the labour market when they fall on hard times. these include providing wage cans par the ablley anc arereas so the as areathe past on ameraca in the poricis farte st pored als\n",
+      "----- diversity: 1.2\n",
+      " keep more americans in the labour market when they fall on hard times. these include providing wage chalrbunisce thag-fatliy ny ducs larforod eno be.\n",
+      "sthean muthe natadotcocon in poidelo a5le that . \n",
+      "epoch: 38 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 1.9023 - val_loss: 2.0436\n",
+      "----- generating with seed: there are rules to guard against systemic failure and ensure fair competition.\n",
+      "\n",
+      "post-crisis reforms \n",
+      "----- diversity: 0.5\n",
+      "there are rules to guard against systemic failure and ensure fair competition.\n",
+      "\n",
+      "post-crisis reforms the workess congest portens the grewting tome sucting the peinco soysed on the pesteres for the to m\n",
+      "----- diversity: 1.2\n",
+      "there are rules to guard against systemic failure and ensure fair competition.\n",
+      "\n",
+      "post-crisis reforms we he chat in wragee esurco boollon. pitten for fore sy will, wive wadl iad rcoe fhall cnorstatsorao\n",
+      "epoch: 39 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 119s - loss: 1.8795 - val_loss: 2.0364\n",
+      "----- generating with seed: d actually make the problem worse rather than better, it is important to remember that capitalism ha\n",
+      "----- diversity: 0.5\n",
+      "d actually make the problem worse rather than better, it is important to remember that capitalism has ence more beterens to mes merica in leve the peanst are rade on thiin wort enor hat fingre the for\n",
+      "----- diversity: 1.2\n",
+      "d actually make the problem worse rather than better, it is important to remember that capitalism hagounirg fumcratiny ecances grseogeconvelyy b-eofer theakfom butibrsstecofre, arcls, phevtingr nthiti\n",
+      "epoch: 40 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 119s - loss: 1.8694 - val_loss: 2.0404\n",
+      "----- generating with seed: to a past that is not possible to restoreand that, for most americans, never existed at all? \n",
+      "\n",
+      "its t\n",
+      "----- diversity: 0.5\n",
+      "to a past that is not possible to restoreand that, for most americans, never existed at all? \n",
+      "\n",
+      "its the the age work and semarisal on the porepar workers op of the for the prowect ane the to and in our\n",
+      "----- diversity: 1.2\n",
+      "to a past that is not possible to restoreand that, for most americans, never existed at all? \n",
+      "\n",
+      "its thp ineq aut y1ve ame ricar thaly neursod ensured anne pruutic1. ins non manvinb seecte pasl y.\n",
+      "lubs \n",
+      "epoch: 41 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 118s - loss: 1.8481 - val_loss: 2.0377\n",
+      "----- generating with seed: o high-quality child care and early learning, would add flexibility for employees and employers. ref\n",
+      "----- diversity: 0.5\n",
+      "o high-quality child care and early learning, would add flexibility for employees and employers. refirce the the past a the pronter and that ching-tin co of a bating the to mers reares the recensmer f\n",
+      "----- diversity: 1.2\n",
+      "o high-quality child care and early learning, would add flexibility for employees and employers. refontm il vest fecinott- pore. ts wevi vendefal corna a-gisperpiosty tece ey ahe col sue uur a vutitre\n",
+      "epoch: 42 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 118s - loss: 1.8349 - val_loss: 2.0324\n",
+      "----- generating with seed: average tax rates on households projected to earn over 8m per yearthe top 0.1by nearly 7 percentage \n",
+      "----- diversity: 0.5\n",
+      "average tax rates on households projected to earn over 8m per yearthe top 0.1by nearly 7 percentage the port mere the fingreats the pardeng the arcous and the preco the a sonthere we the edonecis the \n",
+      "----- diversity: 1.2\n",
+      "average tax rates on households projected to earn over 8m per yearthe top 0.1by nearly 7 percentage und sonte.\n",
+      "\n",
+      "btiby fimess boutexilw atiinla is of iceoneicad halkest jatigrparinor have merest onslri\n",
+      "epoch: 43 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 119s - loss: 1.8094 - val_loss: 2.0400\n",
+      "----- generating with seed: en decades in the making. while i am proud of what my administration has accomplished these past eig\n",
+      "----- diversity: 0.5\n",
+      "en decades in the making. while i am proud of what my administration has accomplished these past eigniste the more and resente the and more the mame of the ans inderiss to the exporting the ally the p\n",
+      "----- diversity: 1.2\n",
+      "en decades in the making. while i am proud of what my administration has accomplished these past eigne werl aod iblllongsif lanakifis no ujveve balanpanilizatisg uprels aleure runctwony ingyescaw coum\n",
+      "epoch: 44 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 116s - loss: 1.7944 - val_loss: 2.0390\n",
+      "----- generating with seed: for workers who cannot get a new job that pays as much as their old one. increasing access to high-q\n",
+      "----- diversity: 0.5\n",
+      "for workers who cannot get a new job that pays as much as their old one. increasing access to high-quality or ther so tome in rate and suplores the wolls and dest on the decont mere oup presingenin de\n",
+      "----- diversity: 1.2\n",
+      "for workers who cannot get a new job that pays as much as their old one. increasing access to high-qoaling comprisscorable the fprofmerns mpiat-pioss bati teer?-tore aconoblfast hove mireboor crasis. \n",
+      "epoch: 45 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 1.7907 - val_loss: 2.0454\n",
+      "----- generating with seed: t years should also give the world some measure of hope. despite all manner of division and discord,\n",
+      "----- diversity: 0.5\n",
+      "t years should also give the world some measure of hope. despite all manner of division and discord, the orem boteden so cens see of anderest proving ald andiciling in whele thes foricad fistarity sov\n",
+      "----- diversity: 1.2\n",
+      "t years should also give the world some measure of hope. despite all manner of division and discord, eriste snablen coplell jalkddinc batero-. ppolkyo4 p oppertuctisg mians, ant deenhe dwating pabtid \n",
+      "epoch: 46 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 116s - loss: 1.7670 - val_loss: 2.0457\n",
+      "----- generating with seed: st precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on\n",
+      "----- diversity: 0.5\n",
+      "st precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on who acronges for the to biting andemints the perecteand aud andisisiss of the sore the infthe be bu\n",
+      "----- diversity: 1.2\n",
+      "st precipitously.\n",
+      "\n",
+      "there are many ways to keep more americans in the labour market when they fall on heplevasith yisnuw kres, wenlor coed gnoghave shate it raste alist wolis potucing\n",
+      "\n",
+      "eoremerses firss\n",
+      "epoch: 47 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 117s - loss: 1.7599 - val_loss: 2.0498\n",
+      "----- generating with seed: people want, regardless of how we divide up the pie.\n",
+      "\n",
+      "a major source of the recent productivity slow\n",
+      "----- diversity: 0.5\n",
+      "people want, regardless of how we divide up the pie.\n",
+      "\n",
+      "a major source of the recent productivity slowhit vere the growe the greas fored the amprice tore wath a destatilisiss and partion of the parica s\n",
+      "----- diversity: 1.2\n",
+      "people want, regardless of how we divide up the pie.\n",
+      "\n",
+      "a major source of the recent productivity slowint houdd, sotsapls the edep thith detpiainon toplorgriin  off coudepikiond micars focted, whor hove\n",
+      "epoch: 48 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 40594s - loss: 1.7516 - val_loss: 2.0402\n",
+      "----- generating with seed: aradox that defines our world today. the world is more prosperous than ever before and yet our socie\n",
+      "----- diversity: 0.5\n",
+      "aradox that defines our world today. the world is more prosperous than ever before and yet our sociess and the proincess as the thore grow elpore to deas coul shald has decentite and ay the to ex mere\n",
+      "----- diversity: 1.2\n",
+      "aradox that defines our world today. the world is more prosperous than ever before and yet our socies, matifen mingel mard robe, adiky mafe hives ecanencase hag conding p impitevicived thoec 2enesomab\n",
+      "epoch: 49 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 121s - loss: 1.7251 - val_loss: 2.0380\n",
+      "----- generating with seed: ants.\n",
+      "\n",
+      "the results are clear: a more durable, growing economy; 15m new private-sector jobs since ear\n",
+      "----- diversity: 0.5\n",
+      "ants.\n",
+      "\n",
+      "the results are clear: a more durable, growing economy; 15m new private-sector jobs since eare of the prowtho forten gronge more the butiec anderty and ploverty at of the losg the the prolinge \n",
+      "----- diversity: 1.2\n",
+      "ants.\n",
+      "\n",
+      "the results are clear: a more durable, growing economy; 15m new private-sector jobs since eare hage. tat co5m 8ass endtyans, ans,erd-tevneoblod, icoronced somploster mogivel sequaling yhorlyd m\n",
+      "epoch: 50 / 50\n",
+      "Train on 14569 samples, validate on 3643 samples\n",
+      "Epoch 1/1\n",
+      "14569/14569 [==============================] - 122s - loss: 1.7119 - val_loss: 2.0483\n",
+      "----- generating with seed:  the auto industry rescued. i enacted a larger and more front-loaded fiscal stimulus than even presi\n",
+      "----- diversity: 0.5\n",
+      " the auto industry rescued. i enacted a larger and more front-loaded fiscal stimulus than even presivingite the eas cofsens ef more with abl in ofrend and in ans to reat rese meconomy are of the me fo\n",
+      "----- diversity: 1.2\n",
+      " the auto industry rescued. i enacted a larger and more front-loaded fiscal stimulus than even presiarininas foc bugeuted ba loste. m0ngit 200y avss roucr weec ans reseksond fion anl prentserngdes, bc\n"
+     ]
+    }
+   ],
+   "source": [
+    "epochs = 50\n",
+    "prediction_length = 100\n",
+    "\n",
+    "for iteration in range(epochs):\n",
+    "    \n",
+    "    print 'epoch:', iteration + 1, '/', epochs\n",
+    "    model.fit(X, y, validation_split=0.2, batch_size=256, nb_epoch=1, callbacks=callbacks_list)\n",
+    "    \n",
+    "    # get random starting point for seed\n",
+    "    start_index = random.randint(0, len(raw_text) - seq_length - 1)\n",
+    "    # extract seed sequence from raw text\n",
+    "    seed = raw_text[start_index: start_index + seq_length]\n",
+    "    \n",
+    "    print '----- generating with seed:', seed\n",
+    "    \n",
+    "    for diversity in [0.5, 1.2]:\n",
+    "        generate(seed, prediction_length, diversity)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved data to -basic_data.pickle\n",
+      "Compressed pickle size: 80934860\n"
+     ]
+    }
+   ],
+   "source": [
+    "pickle_file = '-basic_data.pickle'\n",
+    "\n",
+    "try:\n",
+    "    f = open(pickle_file, 'wb')\n",
+    "    save = {\n",
+    "        'X': X,\n",
+    "        'y': y,\n",
+    "        'int_to_char': int_to_char,\n",
+    "        'char_to_int': char_to_int,\n",
+    "    }\n",
+    "    pickle.dump(save, f, pickle.HIGHEST_PROTOCOL)\n",
+    "    f.close()\n",
+    "except Exception as e:\n",
+    "    print 'Unable to save data to', pickle_file, ':', e\n",
+    "    raise\n",
+    "    \n",
+    "statinfo = os.stat(pickle_file)\n",
+    "print 'Saved data to', pickle_file\n",
+    "print 'Compressed pickle size:', statinfo.st_size"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/notebooks/week-6/lab6_3.ipynb b/notebooks/week-6/lab6_3.ipynb
new file mode 100644
index 0000000..3b614bf
--- /dev/null
+++ b/notebooks/week-6/lab6_3.ipynb
@@ -0,0 +1,308 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Using TensorFlow backend.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense\n",
+    "from keras.layers import Dropout\n",
+    "from keras.layers import LSTM\n",
+    "from keras.callbacks import ModelCheckpoint\n",
+    "from keras.utils import np_utils\n",
+    "\n",
+    "from time import gmtime, strftime\n",
+    "import os\n",
+    "import re\n",
+    "import pickle\n",
+    "import random\n",
+    "import sys"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "length of text: 141266\n",
+      "text preview: alices adventures in wonderland\n",
+      "\n",
+      "lewis carroll\n",
+      "\n",
+      "the millennium fulcrum edition 3.0\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "chapter i. down the rabbit-hole\n",
+      "\n",
+      "alice was beginning to get very tired of sitting by her sister on the\n",
+      "bank, and of having nothing to do: once or twice she had peeped into the\n",
+      "book her sister was reading, but it had no pictures or conversations in\n",
+      "it, and what is the use of a book, thought alice without pictures or\n",
+      "conversations?\n",
+      "\n",
+      "so she was considering in her own mind as well as she could, for the\n",
+      "hot day mad\n"
+     ]
+    }
+   ],
+   "source": [
+    "filename = \"data/wonderland.txt\"\n",
+    "raw_text = open(filename).read()\n",
+    "\n",
+    "raw_text = re.sub('[^\\nA-Za-z0-9 ,.:;?!-]+', '', raw_text)\n",
+    "raw_text = raw_text.lower()\n",
+    "\n",
+    "n_chars = len(raw_text)\n",
+    "print \"length of text:\", n_chars\n",
+    "print \"text preview:\", raw_text[:500]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "number of unique characters found: 37\n",
+      "a - maps to -> 11\n",
+      "25 - maps to -> o\n",
+      "Total sequences:  141166\n",
+      "one eye; i seem to see some meaning in them, after all. --said\n",
+      "i could not swim-- you cant swim, can -->  \n",
+      "X dims --> (141166, 100, 37)\n",
+      "y dims --> (141166, 37)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# write your code here\n",
+    "\n",
+    "# extract all unique characters in the text\n",
+    "#The method list() takes sequence types and converts them to lists. This is used to convert a given tuple into list\n",
+    "chars = sorted(list(set(raw_text)))\n",
+    "n_vocab = len(chars)\n",
+    "print \"number of unique characters found:\", n_vocab\n",
+    "\n",
+    "# create mapping of characters to integers and back\n",
+    "#create dictionary of characters and numbers\n",
+    "char_to_int = dict((c, i) for i, c in enumerate(chars))\n",
+    "int_to_char = dict((i, c) for i, c in enumerate(chars))\n",
+    "\n",
+    "# test our mapping\n",
+    "print 'a', \"- maps to ->\", char_to_int[\"a\"]\n",
+    "print 25, \"- maps to ->\", int_to_char[25]\n",
+    "\n",
+    "\n",
+    "#set longer sequence length to let the model trace back more and thus get higher accuracy\n",
+    "seq_length = 100\n",
+    "\n",
+    "inputs = []\n",
+    "outputs = []\n",
+    "\n",
+    "for i in range(0, n_chars - seq_length, 1):\n",
+    "    inputs.append(raw_text[i:i + seq_length])\n",
+    "    outputs.append(raw_text[i + seq_length])\n",
+    "    \n",
+    "n_sequences = len(inputs)\n",
+    "print \"Total sequences: \", n_sequences\n",
+    "\n",
+    "\n",
+    "#shuffle the indecs that are shared by both inputs and outputs \n",
+    "#for convenience of spliting them into training and test data\n",
+    "indeces = range(len(inputs))\n",
+    "random.shuffle(indeces)\n",
+    "\n",
+    "inputs = [inputs[x] for x in indeces]\n",
+    "outputs = [outputs[x] for x in indeces]\n",
+    "print inputs[0], \"-->\", outputs[0]\n",
+    "\n",
+    "\n",
+    "# create two empty numpy array with the proper dimensions\n",
+    "X = np.zeros((n_sequences, seq_length, n_vocab), dtype=np.bool)\n",
+    "y = np.zeros((n_sequences, n_vocab), dtype=np.bool)\n",
+    "\n",
+    "# iterate over the data and build up the X and y data sets\n",
+    "# by setting the appropriate indices to 1 in each one-hot vector\n",
+    "for i, example in enumerate(inputs):\n",
+    "    for t, char in enumerate(example):\n",
+    "        X[i, t, char_to_int[char]] = 1\n",
+    "    y[i, char_to_int[outputs[i]]] = 1\n",
+    "    \n",
+    "print 'X dims -->', X.shape\n",
+    "print 'y dims -->', y.shape\n",
+    "\n",
+    "\n",
+    "# define the LSTM model\n",
+    "model = Sequential()\n",
+    "\n",
+    "model.add(LSTM(32, return_sequences=True, input_shape=(X.shape[1], X.shape[2])))\n",
+    "\n",
+    "#add up the dropout probability to minimize the overfitting problem\n",
+    "model.add(Dropout(0.30))\n",
+    "\n",
+    "#add extra recurrent layer to train the model with more complex structures\n",
+    "#reducing the hidden neurons in Recurrent layers to decay the model complexity and speed up learning process\n",
+    "model.add(LSTM(64))\n",
+    "#add up the dropout probability to minimize the overfitting problem\n",
+    "model.add(Dropout(0.30))\n",
+    "\n",
+    "model.add(Dense(y.shape[1], activation='softmax'))\n",
+    "\n",
+    "model.compile(loss='categorical_crossentropy', optimizer='adam')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def sample(preds, temperature=1.0):\n",
+    "    preds = np.asarray(preds).astype('float64')\n",
+    "    preds = np.log(preds) / temperature\n",
+    "    exp_preds = np.exp(preds)\n",
+    "    preds = exp_preds / np.sum(exp_preds)\n",
+    "    probas = np.random.multinomial(1, preds, 1)\n",
+    "    return np.argmax(probas)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def generate(sentence, sample_length=50, diversity=0.35):\n",
+    "    generated = sentence\n",
+    "    sys.stdout.write(generated)\n",
+    "\n",
+    "    for i in range(sample_length):\n",
+    "        x = np.zeros((1, X.shape[1], X.shape[2]))\n",
+    "        for t, char in enumerate(sentence):\n",
+    "            x[0, t, char_to_int[char]] = 1.\n",
+    "\n",
+    "        preds = model.predict(x, verbose=0)[0]\n",
+    "        next_index = sample(preds, diversity)\n",
+    "        next_char = int_to_char[next_index]\n",
+    "\n",
+    "        generated += next_char\n",
+    "        sentence = sentence[1:] + next_char\n",
+    "\n",
+    "        sys.stdout.write(next_char)\n",
+    "        sys.stdout.flush()\n",
+    "    print"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "filepath=\"-Alice_LSTM.hdf5\"\n",
+    "checkpoint = ModelCheckpoint(filepath, monitor='loss', verbose=0, save_best_only=True, mode='min')\n",
+    "callbacks_list = [checkpoint]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "epoch: 1 / 50\n",
+      "Train on 112932 samples, validate on 28234 samples\n",
+      "Epoch 1/1\n"
+     ]
+    }
+   ],
+   "source": [
+    "epochs = 50\n",
+    "prediction_length = 100\n",
+    "\n",
+    "for iteration in range(epochs):\n",
+    "    \n",
+    "    print 'epoch:', iteration + 1, '/', epochs\n",
+    "    model.fit(X, y, validation_split=0.2, batch_size=256, nb_epoch=1, callbacks=callbacks_list)\n",
+    "    \n",
+    "    # get random starting point for seed\n",
+    "    start_index = random.randint(0, len(raw_text) - seq_length - 1)\n",
+    "    # extract seed sequence from raw text\n",
+    "    seed = raw_text[start_index: start_index + seq_length]\n",
+    "    \n",
+    "    print '----- generating with seed:', seed\n",
+    "    \n",
+    "    for diversity in [0.5, 1.2]:\n",
+    "        generate(seed, prediction_length, diversity)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "anaconda-cloud": {},
+  "kernelspec": {
+   "display_name": "Python [default]",
+   "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": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}