ex4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Networks Learning\n",
    "\n",
    "# Introduction\n",
    "\n",
    "The backpropagation algorithm will be implemented for neural networks and it will be applied to the task of hand-written digit recognition.\n",
    "\n",
    "# Neural Networks\n",
    "\n",
    "The backpropagation algorithm will be implemented to learn the parameters for the neural network.\n",
    "\n",
    "## Visualizing the Data\n",
    "\n",
    "Load the data and display it on a 2-dimensional plot by calling the function `displayData`.\n",
    "\n",
    "There are 5000 training examples in `ex3data1.mat`, where each training example is a 20 pixel by 20 pixel grayscale image of the digit. Each pixel is represented by a floating point number indicating the grayscale intensity at that location. The 20 by 20 grid of pixels is “unrolled” into a 400-dimensional vector. Each of these training examples becomes a single row in the data matrix $X$. This gives a 5000 by 400 matrix $X$ where every row is a training example for a handwritten digit image.\n",
    "\n",
    "$$X=\\begin{bmatrix}\n",
    "    -(x^{(1)})^T-\\\\\n",
    "    -(x^{(2)})^T-\\\\\n",
    "    \\vdots\\\\\n",
    "    -(x^{(m)})^T-\n",
    "\\end{bmatrix}$$\n",
    "\n",
    "The second part of the training set is a 5000-dimensional vector $y$ that contains labels for the training set. To make things more compatible with Octave/MATLAB indexing, where there is no zero index, the digit zero has been mapped to the value ten. Therefore, a “0” digit is labeled as “10”, while the digits “1” to “9” are labeled as “1” to “9” in their natural order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5000, 400) (5000, 1)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWMAAAFbCAYAAAAA6+8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nNDDbJoopAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111377e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np \n",
    "from scipy.io import loadmat\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "data = loadmat('ex3data1.mat')\n",
    "X = data['X']\n",
    "y = data['y']\n",
    "\n",
    "print(X.shape, y.shape)\n",
    "\n",
    "def displayData(X):\n",
    "    \"\"\"\n",
    "    Select randomly 100 rows from X,\n",
    "    plot them as 20x20 pixel grayscale images,\n",
    "    and combine them to one figure with all\n",
    "    100 digits.\n",
    "    \"\"\"\n",
    "    # Create 100 subplots and remove the gaps\n",
    "    # between the subplots with gridspec.\n",
    "    fig, axarr = plt.subplots(10, 10,\n",
    "                             figsize=(6, 6),\n",
    "                             gridspec_kw={'wspace':0,\n",
    "                                          'hspace':0})\n",
    "    for i in range(10):\n",
    "        for j in range(10):\n",
    "            # Select random indices.\n",
    "            idx = np.random.choice(len(X), 1)\n",
    "            # Index the array X with the indices.\n",
    "            pixels = X[idx] # shape(1, 400)\n",
    "            pixels = pixels.reshape(-1, 20) # shape(20, 20)\n",
    "            axarr[i,j].imshow(pixels.T, cmap='gray_r')\n",
    "            # Remove ticks.\n",
    "            axarr[i,j].set_xticks([])\n",
    "            axarr[i,j].set_yticks([])\n",
    "            # Turn off axes.\n",
    "            axarr[i,j].axis('off')\n",
    "    plt.show()\n",
    "\n",
    "displayData(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Representation\n",
    "\n",
    "The neural network has 3 layers - an input layer, a hidden layer and an output layer. Recall that the inputs are pixel values of digit images. Since the images are of size $20x20$, this gives 400 input layer units (not counting the extra bias unit which always outputs +1). The training data will be loaded into the variables $X$ and $y$.\n",
    "\n",
    "A set of already trained network parameters $(\\Theta^{(1)}, \\Theta^{(2)})$ stored in `ex3weights.mat` will be loaded by `loadmat` into `Theta1` and `Theta2`. The parameters have dimensions that are sized for a neural network with 25 units in the second layer and 10 output units (corresponding to the 10 digit classes)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of Theta1:  (25, 401)\n",
      "Shape of Theta2:  (10, 26)\n",
      "Shape of nn_params:  (10285,)\n"
     ]
    }
   ],
   "source": [
    "# Initialize the parameters.\n",
    "# 20x20 Input images of digits\n",
    "input_layer_size  = 400\n",
    "\n",
    "# 25 Hidden units\n",
    "hidden_layer_size = 25\n",
    "\n",
    "# 10 Labels, from 1 to 10\n",
    "# Note that \"0\" has been mapped to \"10\".\n",
    "num_labels = 10\n",
    "\n",
    "# Load the weights into variables Theta1 and Theta2.\n",
    "weights = loadmat('ex3weights.mat')\n",
    "Theta1 = weights['Theta1']\n",
    "Theta2 = weights['Theta2']\n",
    "\n",
    "# Create a list of my Thetas.\n",
    "Thetas = [Theta1, Theta2]\n",
    "\n",
    "# Unroll parameters and then merge/concatenate.\n",
    "unrolled_Thetas = [Thetas[i].ravel() for i,_ in enumerate(Thetas)]\n",
    "nn_params = np.concatenate(unrolled_Thetas)\n",
    "\n",
    "print(\"Shape of Theta1: \", Theta1.shape)\n",
    "print(\"Shape of Theta2: \", Theta2.shape)\n",
    "print(\"Shape of nn_params: \", nn_params.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 1 depicts a simple neural network with 3 layers. The first input layer has three units, the input units $x_i$ and is expressed as:\n",
    "\n",
    "$$x=\\alpha^{(1)}=(\\alpha_1^{(1)},\\alpha_2^{(1)},\\alpha_3^{(1)})^T$$\n",
    "\n",
    "The second hidden layer is similarly:\n",
    "\n",
    "$$\\alpha^{(2)}=(\\alpha_1^{(2)},\\alpha_2^{(2)},\\alpha_3^{(2)})^T$$\n",
    "\n",
    "where\n",
    "\n",
    "$$\\alpha_1^{(2)}=g(\\Theta_{11}^{(1)}\\alpha_1^{(1)}+\\Theta_{12}^{(1)}\\alpha_2^{(1)}+\\Theta_{13}^{(1)}\\alpha_3^{(1)})$$\n",
    "$$\\alpha_2^{(2)}=g(\\Theta_{21}^{(1)}\\alpha_1^{(1)}+\\Theta_{22}^{(1)}\\alpha_2^{(1)}+\\Theta_{23}^{(1)}\\alpha_3^{(1)})$$\n",
    "$$\\alpha_3^{(2)}=g(\\Theta_{31}^{(1)}\\alpha_1^{(1)}+\\Theta_{32}^{(1)}\\alpha_2^{(1)}+\\Theta_{33}^{(1)}\\alpha_3^{(1)})$$\n",
    "\n",
    "and $g(z) = \\frac{1}{1 + e^{-z}}$ is the logistic function.\n",
    "\n",
    "\n",
    "The third output layer has only one unit, which returns the hypothesis function:\n",
    "\n",
    "$$h_{\\Theta}(x)=\\alpha^{(3)}=(\\alpha_1^{(3)})=g(\\Theta_{11}^{(2)}\\alpha_1^{(2)}+\\Theta_{12}^{(2)}\\alpha_2^{(2)}+\\Theta_{13}^{(2)}\\alpha_3^{(2)})$$\n",
    "\n",
    "![Figure 1](simple_nn.png \"Neural Network Model\")\n",
    "\n",
    "## Feedforward and Cost Function\n",
    "\n",
    "First the cost function and gradient for the neural network will be implemented. Create a `nnCostFunction` fuction to return the cost. Recall that the cost function for the neural network (without regularization) is\n",
    "\n",
    "$$J(\\theta)=\\frac{1}{m}\\sum_{i=1}^m\\sum_{k=1}^K[-y_k^{(i)}log((h_\\theta(x^{(i)})_k)-(1-y_k^{(i)})log((1-h_\\theta (x^{(i)}))_k)]$$\n",
    "\n",
    "Note that $h_{\\theta}(x^{(i)})_k=\\alpha_k^{(3)}$ is the activation (output value) of the $k^{th}$ output unit. Also, recall that whereas the original labels (in the variable $y$) were 1, 2, ..., 10, for the purpose of training a neural network, the labels should be recoded as vectors containing only values 0 or 1, so that\n",
    "\n",
    "$$y=\\begin{bmatrix}\n",
    "    1\\\\\n",
    "    0\\\\\n",
    "    0\\\\\n",
    "    \\vdots\\\\\n",
    "    0\\\\\n",
    "\\end{bmatrix},\\hspace{0.5cm}y=\\begin{bmatrix}\n",
    "    0\\\\\n",
    "    1\\\\\n",
    "    0\\\\\n",
    "    \\vdots\\\\\n",
    "    0\\\\\n",
    "\\end{bmatrix},...\\hspace{0.5cm}or\\hspace{0.5cm}\n",
    "y=\\begin{bmatrix}\n",
    "    0\\\\\n",
    "    0\\\\\n",
    "    0\\\\\n",
    "    \\vdots\\\\\n",
    "    1\\\\\n",
    "\\end{bmatrix}$$\n",
    "\n",
    "For example, if $x^{(i)}$ is an image of the digit 5, then the corresponding $y^{(i)}$ should be a 10-dimensional vector with $y_5=1$, and the other elements equal to 0.\n",
    "\n",
    "The feedforward computation that computes $h_{\\theta}(x^{(i)})$ should be implemented for every example $i$ and the cost over all examples should be summed. The code should also work for a dataset of any size, with any number of labels assuming that there are always at least $K≥3$ labels.\n",
    "\n",
    "** Note **\n",
    "\n",
    "The matrix $X$ contains the examples in rows (i.e., $X(i,:)$' is the i-th training example $x^{(i)}$, expressed as a $nx1$ vector). To complete the code in `nnCostFunction` function, we need to add the column of $1$'s to the $X$ matrix. The parameters for each unit in the neural network are represented in `Theta1` and `Theta2` as one row. Speciffically, the first row of Theta1 corresponds to the first hidden unit in the second layer. We can use a for-loop over the examples to compute the cost.\n",
    "\n",
    "## Regularized Cost Function\n",
    "\n",
    "For neural networks, the cost function with regularization is going to be slightly more complicated:\n",
    "\n",
    "$$J(\\theta)=\\frac{1}{m}\\sum_{i=1}^m\\sum_{k=1}^K[-y_k^{(i)}log((h_\\theta(x^{(i)})_k)-(1-y_k^{(i)})log((1-h_\\theta (x^{(i)}))_k)]+\\frac{\\lambda}{2m}\\sum_{l=1}^{L-1}\\sum_{i=1}^{s_l}\\sum_{j=1}^{s_{l+1}}(\\Theta_{j,i}^{(1)})^2$$\n",
    "\n",
    "For the particular neural network with the 3 layers, the cost function is given by\n",
    "\n",
    "$$J(\\theta)=\\frac{1}{m}\\sum_{i=1}^m\\sum_{k=1}^K[-y_k^{(i)}log((h_\\theta(x^{(i)})_k)-(1-y_k^{(i)})log((1-h_\\theta (x^{(i)}))_k)]+\\frac{\\lambda}{2m}[\\sum_{j=1}^{25}\\sum_{k=1}^{400}(\\Theta_{j,k}^{(1)})^2+\\sum_{j=1}^{10}\\sum_{k=1}^{25}(\\Theta_{j,k}^{(2)})^2]$$\n",
    "\n",
    "We can assume that the neural network will only have 3 layers - an input layer, a hidden layer and an output layer. However, the code can work for any number of input units, hidden units and outputs units. While we have explicitly listed the indices above for $\\Theta^{(1)}$ and $\\Theta^{(2)}$ for clarity, do note that **the code can in general work with $\\Theta^{(1)}$ and $\\Theta^{(2)}$ of any size.**\n",
    "\n",
    "Note that we should not be regularizing the terms that correspond to the bias. For the matrices `Theta1` and `Theta2`, this corresponds to the first column of each matrix. Calling `nnCostFunction` using the loaded set of parameters for `Theta1` and `Theta2`, and $\\lambda=1$ we should see that the cost is about 0.383770.\n",
    "\n",
    "# Backpropagation\n",
    "\n",
    "To implement the backpropagation algorithm we have to compute the gradient for the neural network cost function. We need to complete the `nnCostFunction` so that it returns an appropriate value for grad. Once we have computed the gradient, we will be able to train the neural network by minimizing the cost function $J(\\Theta)$ using an advanced optimizer from `scipy` such as the `minimize` function.\n",
    "\n",
    "## Sigmoid Gradient\n",
    "\n",
    "Before the implementation of backpropagation, we need first to implement the sigmoid gradient function. The gradient for the sigmoid function can be computed as\n",
    "\n",
    "$$g'(z)=\\frac{d}{dz}g(z)=g(z)(1-g(z))$$\n",
    "\n",
    "where\n",
    "\n",
    "$$sigmoid(z)=g(z)=\\frac{1}{1+e^{-z}}$$\n",
    "\n",
    "When we are done, we can try testing a few values by calling `sigmoidGradient(z)`. For large values (both positive and negative) of $z$, the gradient should be close to $0$. When $z=0$, the gradient should be exactly $0.25$. The code can also work with vectors and matrices. For a matrix, the function can perform the sigmoid gradient function on every element.\n",
    "\n",
    "## Random Initialization\n",
    "\n",
    "When training neural networks, it is important to randomly initialize the parameters for symmetry breaking. One effective strategy for random initialization is to randomly select values for $\\Theta^{(l)}$ uniformly in the range $[-\\epsilon_{init},\\epsilon_{init}]$.\n",
    "\n",
    "We should use $\\epsilon_{init}=0.12$. This range of values ensures that the parameters are kept small and makes the learning more efficient. We need to implement a function `randInitializeWeights` that will initialize the weights for $\\Theta$.\n",
    "\n",
    "**Note** One effective strategy for choosing $\\epsilon_{init}$ is to base it on the number of units in the network. A good choice of $\\epsilon_{init}$ is\n",
    "\n",
    "$$\\epsilon_{init}=\\frac{\\sqrt{6}}{\\sqrt{L_{in}+L_{out}}}$$\n",
    "\n",
    "where $L_{in}=s_l$ and $L_{out}=s_{l}+1$ are the number of units in the layers adjacent to $\\Theta^{(l)}$.\n",
    "\n",
    "## Backpropagation\n",
    "\n",
    "Now, we can implement the backpropagation algorithm. Recall that the intuition behind the backpropagation algorithm is as follows. Given a training example $(x^{(t)},y^{(t)})$, we will first run a \"forward pass\" to compute all the activations throughout the network, including the output value of the hypothesis $h_{\\theta}(x)$. Then, for each node $j$ in layer $l$, we would like to compute an \"error term\" $\\delta_j^{(l)}$ that measures how much that node was \"responsible\" for any errors in our output.\n",
    "\n",
    "For an output node, we can directly measure the difference between the network's activation and the true target value, and use that to define $\\delta_j^{(3)}$ (since layer $3$ is the output layer). For the hidden units, we will compute $\\delta_j^{(l)}$ based on a weighted average of the error terms of the nodes in layer $(l+1)$.\n",
    "\n",
    "In detail, the backpropagation algorithm is shown in Figure 2. To accomplish this we have to implement steps 1 to 4 in a loop that processes one example at a time. Concretely, we should implement a `for-loop` for $t=1:m$ and place steps 1-4 below inside the for-loop, with the $t^{th}$ iteration performing the calculation on the $t^{th}$ training example $(x^{(t)},y^{(t)})$. Step 5 will divide the accumulated gradients by $m$ to obtain the gradients for the neural network cost function.\n",
    "\n",
    "![Figure 2](backprop.png \"Backpropagation Algorithm\")\n",
    "\n",
    "1. Set the input layer's values $(a^{(1)})$ to the $t^{th}$ training example $x^{(t)}$. Perform a feedforward pass (see Figure 3), computing the activations $(z^{(2)},\\alpha^{(2)},z^{(3)}, \\alpha^{(3)})$ for layers $2$ and $3$. Note that we need to add a $+1$ term to ensure that the vectors of activations for layers $\\alpha^{(1)}$ and $\\alpha^{(2)}$ also include the bias unit. ![Figure 3](nnModel.png \"Feedforward Pass\")\n",
    "\n",
    "2. For each output unit $k$ in layer $3$ (the output layer), set $$\\delta_k^{(3)}=(\\alpha_k^{(3)}-y_k)$$ where $y_k\\in{\\{0,1\\}}$ indicates whether the current training example belongs to class $k(y_k=1)$, or if it belongs to a different class $(y_k=0)$.\n",
    "\n",
    "3. For the hidden layer $l=2$, set $$\\delta^{(2)}=(\\Theta^{(2)})^T\\delta^{(3)}.*g'(z^{(2)})$$ Intuitively, $\\delta_j^{(l)}$ is the \"error\" for $\\alpha_j^{(l)}$ (unit $j$ in layer $l$). More formally, the delta values are actually the derivative of the cost function, $\\delta_j^{(l)}=\\frac{\\partial}{\\partial{z_j^{(l)}}}cost(t)$. Recall that our derivative is the slope of a line tangent to the cost function, so the steeper the slope the more incorrect we are.\n",
    "\n",
    "4. Accumulate the gradient from this example using the following formula. Note that we should skip or remove $\\delta_0^{(2)}$. $$\\Delta^{(l)}=\\Delta^{(l)}+\\delta^{(l+1)}(\\alpha^{(l)})^T$$\n",
    "\n",
    "5. Obtain the gradient for the neural network (with regularization) cost function by dividing the accumulated gradients by $\\frac{1}{m}$: $$\\frac{\\partial}{\\partial{\\Theta_{ij}^{(l)}}}J(\\Theta)=D_{ij}^{(l)}=\\frac{1}{m}\\Delta_{ij}^{(l)}\\hspace{20pt}for\\hspace{20pt}j=0$$ $$\\frac{\\partial}{\\partial{\\Theta_{ij}^{(l)}}}J(\\Theta)=D_{ij}^{(l)}=\\frac{1}{m}\\Delta_{ij}^{(l)}+\\frac{\\lambda}{m}\\Theta_{ij}^{(l)}\\hspace{20pt}for\\hspace{20pt}j\\geq1$$ Note that we should not be regularizing the first column of $\\Theta^{(l)}$ which is used for the bias term. Furthermore, in the parameters $\\Theta_{ij}^{(l)}$, $i$ is indexed starting from $1$, and $j$ is indexed starting from $0$. Thus, $$\\Theta^{(1)}=\\begin{bmatrix}\\Theta_{1,0}^{(i)}&\\Theta_{1,1}^{(l)}&\\dots\\\\\\Theta_{2,0}^{(i)}&\\Theta_{2,1}^{(l)}&\\dots\\\\\\vdots&\\vdots&\\ddots\\\\\\end{bmatrix}$$\n",
    "\n",
    "After we have implemented the backpropagation algorithm, we should implement gradient checking. The gradient checking will allow us to increase our confidence that our code is computing the gradients correctly. If the code is correct, we should expect to see a relative difference that is less than `1e-9`.\n",
    "\n",
    "### Summarizing the Algorithms\n",
    "\n",
    "In general, the vectorized algorithms to be applied are:\n",
    "\n",
    "**Forward Propagation**\n",
    "\n",
    "1. $\\alpha^{(1)}=x$\n",
    "2. $\\text{for }l=1,...,L-1:$\n",
    "3. $\\quad z^{(l+1)}=\\Theta^{(l)}a^{(l)}$\n",
    "4. $\\quad \\alpha^{(l+1)}=g(z^{(l+1)})$\n",
    "5. $\\text{end}$\n",
    "6. $h_{\\Theta}(x)=\\alpha^{(L)}$\n",
    "\n",
    "**Back Propagation**\n",
    "\n",
    "1. $\\delta^{(L)}=a^{(L)}-y$\n",
    "2. $\\text{for } l= L-1,...,2:$\n",
    "3. $\\quad \\delta^{(l)}=(\\Theta^{(l)})^T\\delta^{(l + 1)}.*g'(z^{(l)})$\n",
    "4. $\\text{end}$\n",
    "\n",
    "**Algorithm for the gradient of $J(\\Theta)$**\n",
    "\n",
    "1. $\\text{Set } \\Delta^{(l)}:=0, \\text{ for } l=1, ..., L-1$\n",
    "2. $\\text{For training example } t=1, ..., m:$\n",
    "3. $\\quad \\text{Set } \\alpha^{(1)}=x^{(t)}$\n",
    "4. $\\quad \\text{Perform forward propagation to compute } \\alpha^{(l)} \\text{ for } l=1,...,L$\n",
    "5. $\\quad \\text{Using } y^{(t)}, \\text{ compute } \\delta^{(L)}=\\alpha^{(L)}-y^{(t)}$\n",
    "6. $\\quad \\text{Perform back propagation computing } \\delta^{(L-1)},\\delta^{(L-2)},...,\\delta^{(2)} \\text{ using } \\delta^{(l)}=(\\Theta^{(l)})^T\\delta^{(l + 1)}.*g'(z^{(l)}) \\text{ where } g'(z^{(l)})=\\alpha^{(l)}.*(1-\\alpha^{(l)})$\n",
    "6. $\\quad \\Delta^{(l)}:=\\Delta^{(l)}+\\delta^{(l+1)}(a^{(l)})^T \\text{ for } l=L-1, ..., 1$\n",
    "7. $\\text{end}$\n",
    "\n",
    "Hence we update our new $\\Delta$ matrix.\n",
    "\n",
    "$D_{ij}^{(l)}:=\\frac{1}{m}\\Delta_{ij}^{(l)}\\text{ for }j=0$\n",
    "\n",
    "$D_{ij}^{(l)}:=\\frac{1}{m}(\\Delta_{ij}^{(l)}+\\lambda \\Theta_{ij}^{(l)})\\text{ for }j\\geq1$\n",
    "\n",
    "The capital-delta matrix $D$ is used as an \"accumulator\" to add up our values as we go along and eventually compute our partial derivative, $\\frac{\\partial}{\\partial{\\Theta_{ij}^{(l)}}}J(\\Theta)=D_{ij}^{(l)}$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.38376985909092348,\n",
       " array([  6.18712766e-05,  -2.11248326e-12,   4.38829369e-13, ...,\n",
       "          4.70513145e-05,  -5.01718610e-04,   5.07825789e-04]))"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the logistic sigmoid function.\n",
    "def sigmoid(z):\n",
    "    \"\"\"\n",
    "    Computes the sigmoid function.\n",
    "    Args:\n",
    "        z: float, vector, matrix\n",
    "    Returns:\n",
    "        sigmoid: float, vector, matrix\n",
    "    \"\"\"\n",
    "    sigmoid = 1 / (1 + np.exp(-z))\n",
    "    return sigmoid\n",
    "\n",
    "# Create the sigmoid gradient function.\n",
    "def sigmoidGradient(z):\n",
    "    \"\"\"\n",
    "    Computes the gradient of the sigmoid function\n",
    "    evaluated at z. This works regardless if z is\n",
    "    a matrix or a vector. In particular, if z is\n",
    "    a vector or matrix, it returns the gradient for\n",
    "    each element.\n",
    "    Args:\n",
    "        z: float, vector, matrix\n",
    "    Returns:\n",
    "        g_prime: float, vector, matrix\n",
    "    \"\"\"\n",
    "    g_prime = sigmoid(z) * (1 - sigmoid(z))\n",
    "    return g_prime\n",
    "\n",
    "# Create the neural network cost function.\n",
    "def nnCostFunction(nn_params,\n",
    "                   input_layer_size,\n",
    "                   hidden_layer_size,\n",
    "                   num_labels,\n",
    "                   X, y, lambda_coef):\n",
    "    \"\"\"\n",
    "    Implements the neural network cost function for a two layer\n",
    "    neural network which performs classification.\n",
    "    Computes the cost and gradient of the neural network. The\n",
    "    parameters for the neural network are \"unrolled\" into the vector\n",
    "    nn_params and are converted back into the weight matrices.\n",
    "    The returned parameter grad is an \"unrolled\" vector of the\n",
    "    partial derivatives of the neural network.\n",
    "    Args:\n",
    "        nn_params: vector array\n",
    "        input_layer_size: int\n",
    "        hidden_layer_size: int\n",
    "        num_labels: int\n",
    "        X: array (m, input_layer_size)\n",
    "        y: array (m, num_labels)\n",
    "        lambda_coef: int\n",
    "    Returns:\n",
    "        J: float\n",
    "        grad: vector array\n",
    "    \"\"\"\n",
    "    # Reshape nn_params back into the parameters Theta1 and Theta2,\n",
    "    # the weight matrices for our 2 layer neural network.\n",
    "    Theta1 = np.reshape(nn_params[:hidden_layer_size * (input_layer_size + 1)],\n",
    "                        (hidden_layer_size, input_layer_size + 1)) # (25,401)\n",
    "    Theta2 = np.reshape(nn_params[hidden_layer_size * (input_layer_size + 1):],\n",
    "                        (num_labels, hidden_layer_size + 1)) # (10,26)\n",
    "\n",
    "    # Get the number of training examples, m.\n",
    "    m = X.shape[0]\n",
    "    \n",
    "    # Insert a 1's column for the bias unit.\n",
    "    X = np.insert(X, 0, 1, axis=1) # (5000,401)\n",
    "    \n",
    "    # Perform forward propagation to compute a(l) for l=1,...,L.\n",
    "    # z(l+1) = theta(l)a(l) and a(l+1) = g(z(l+1)).\n",
    "    z2 = np.dot(X, Theta1.T) # (5000, 25)\n",
    "    a2 = sigmoid(z2) # (5000, 25)\n",
    "    \n",
    "    # Add 1's for the bias unit.\n",
    "    a2 = np.insert(a2, 0, 1, axis=1) # (5000,26)\n",
    "    z3 = np.dot(a2, Theta2.T) # (5000, 10)\n",
    "    a3 = sigmoid(z3) # (5000, 10)\n",
    "    \n",
    "    # Create a y matrix of shape (m, K)\n",
    "    # for later use in recoding.\n",
    "    y_recoded = np.zeros((m, num_labels)) # (5000, 10)\n",
    "    \n",
    "    # Initialize Delta matrices.\n",
    "    D1 = np.zeros((hidden_layer_size, input_layer_size + 1)) # (25,401)\n",
    "    D2 = np.zeros((num_labels, hidden_layer_size + 1)) # (10,26)\n",
    "    \n",
    "    #############################################################\n",
    "    ########## Forward Propagation and Cost Computation #########\n",
    "    #############################################################\n",
    "    \n",
    "    # Initialize cost.\n",
    "    j = 0\n",
    "    # Fwd pass; for training example t = 1,...,m:\n",
    "    for t in range(m):\n",
    "        x_t = X[t]\n",
    "        \n",
    "        # Recode the categorical integer values of  y\n",
    "        # as vectors with all values set to zeros except\n",
    "        # for one value set to \"1\", which indicates whether\n",
    "        # it belongs to class k (yk = 1).\n",
    "        y_recoded[t, y[t] - 1] = 1\n",
    "            \n",
    "        # Compute cost for every training example.\n",
    "        j += np.sum(-y_recoded[t] * np.log(a3[t]) - (1 - y_recoded[t]) * np.log(1 - a3[t])) # float\n",
    "\n",
    "        ###############################################################\n",
    "        ########## Back Propagation and Gradients Computation #########\n",
    "        ###############################################################\n",
    "    \n",
    "        # Compute the error delta.\n",
    "        d_3 = a3[t] - y_recoded[t] # (10,)\n",
    "        d_3 = d_3.reshape(-1,1) # (10,1)\n",
    "        \n",
    "        # Perform back propagation.\n",
    "        # In the parameters Thetas_i,j, i are indexed\n",
    "        # starting from 1, and j is indexed starting from  0.\n",
    "        d_2 = np.dot(Theta2.T[1:,:], d_3) * sigmoidGradient(z2[t].reshape(-1,1)) # (25,10)x(10,1).*(25,1) = (25,1)\n",
    "        D1 += np.dot(d_2, x_t.reshape(-1, 1).T) # (25,401)\n",
    "        D2 += np.dot(d_3, a2[t].reshape(1,-1)) # (10,26)\n",
    "\n",
    "    # Compute total cost.\n",
    "    J = j / m # float\n",
    "    \n",
    "    # Compute the regularization term.\n",
    "    # We should not be regularizing the first column of theta,\n",
    "    # which is used for the bias term.\n",
    "    Theta1_sum = np.sum(np.square(Theta1[:,1:])) # float\n",
    "    Theta2_sum = np.sum(np.square(Theta2[:,1:])) # float\n",
    "    reg_term = Theta1_sum + Theta2_sum # float\n",
    "    \n",
    "    # Compute total cost with regularization.\n",
    "    J = J + (lambda_coef / (2 * m)) * reg_term # float\n",
    "    \n",
    "    # Update our new Delta matrices with regularization.\n",
    "    # We should not be regularizing the first column of theta,\n",
    "    # which is used for the bias term.\n",
    "    # First devide every value in Deltas with m.\n",
    "    D1 = D1 / m\n",
    "    D2 = D2 / m\n",
    "    D1[:,1:] = D1[:,1:] + (lambda_coef / m) * Theta1[:,1:]\n",
    "    D2[:,1:] = D2[:,1:] + (lambda_coef / m) * Theta2[:,1:]\n",
    "    \n",
    "    # Unroll gradients.\n",
    "    Deltas = [D1, D2]\n",
    "    unrolled_Deltas = [Deltas[i].ravel() for i,_ in enumerate(Deltas)]\n",
    "    grad = np.concatenate(unrolled_Deltas)\n",
    "\n",
    "    return J, grad\n",
    "\n",
    "# Create the randomly weights-initialization function.\n",
    "def randInitializeWeights(L_in, L_out):\n",
    "    \"\"\"\n",
    "    Randomly initialize the weights of a layer with L_in\n",
    "    incoming connections and L_out outgoing connections.\n",
    "    Note that the column row of W handles the \"bias\" terms.\n",
    "    Args:\n",
    "        L_in: int\n",
    "        L_out: int\n",
    "    Returns:\n",
    "        W: array (L_out, 1 + L_in)\n",
    "    \"\"\"\n",
    "    # Set the range epsilon.\n",
    "    epsilon_init = 0.12\n",
    "    \n",
    "    # Initialize the W matrix.\n",
    "    W = np.zeros((L_out, 1 + L_in))\n",
    "    \n",
    "    # Produce the random values in the range [-epsilon, epsilon].\n",
    "    W = np.random.rand(L_out, 1 + L_in) * (2 * epsilon_init) - epsilon_init\n",
    "    return W\n",
    "\n",
    "nnCostFunction(nn_params,\n",
    "                   input_layer_size,\n",
    "                   hidden_layer_size,\n",
    "                   num_labels,\n",
    "                   X, y, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gradient Checking\n",
    "\n",
    "In our neural network, we are minimizing the cost function $J(\\Theta)$. To perform gradient checking on our parameters, we can imagine “unrolling” the parameters $\\Theta^{(1)}$, $\\Theta^{(2)}$ into a long vector $\\theta$. By doing so, we can think of the cost function being $J(\\theta)$ instead and use the following gradient checking procedure.\n",
    "\n",
    "Suppose we have a function $f_i(\\theta)$ that purportedly computes $\\frac{\\partial}{\\partial{\\theta_i}}J(\\theta)$; We would like to check if $f_i$ is outputting correct derivative values.\n",
    "\n",
    "$$Let\\hspace{10pt}\\theta^{(i+)}=\\theta+\\begin{bmatrix}\n",
    "    0\\\\\n",
    "    0\\\\\n",
    "    \\vdots\\\\\n",
    "    \\epsilon\\\\\n",
    "    \\vdots\\\\\n",
    "    0\\\\\n",
    "\\end{bmatrix}\\hspace{10pt}and\\hspace{10pt}\\theta^{(i-)}=\\theta-\\begin{bmatrix}\n",
    "    0\\\\\n",
    "    0\\\\\n",
    "    \\vdots\\\\\n",
    "    \\epsilon\\\\\n",
    "    \\vdots\\\\\n",
    "    0\\\\\\end{bmatrix}$$\n",
    "\n",
    "So, $\\theta^{(i+)}$ is the same as $\\theta$, except its $i^{th}$ element has been incremented by $\\epsilon$. Similarly, $\\theta^{(i-)}$ is the corresponding vector with the $i^{th}$ element decreased by $\\epsilon$. We can now numerically verify $f_i(\\theta)$’s correctness by checking, for each $i$, that:\n",
    "\n",
    "$$f_i(\\theta)\\approx\\frac{J(\\theta^{(i+)})-J(\\theta^{(i-)})}{2\\epsilon}$$\n",
    "\n",
    "The degree to which these two values should approximate each other will depend on the details of $J$. But assuming $\\epsilon=10^{−4}$, we’ll usually find that the left- and right-hand sides of the above will agree to at least 4 significant digits (and often many more).\n",
    "\n",
    "The implementation of the computation of the numerical gradient is achieved with the `computeNumericalGradient` function.\n",
    "\n",
    "Next, we can run the function `checkNNGradients` which will create a small neural network and dataset that will be used for checking our gradients. If the backpropagation implementation is correct, we should see a relative difference that is less than $10^{-9}$.\n",
    "\n",
    "**Tips:**\n",
    "\n",
    "1. When performing gradient checking, it is much more efficient to use a small neural network with a relatively small number of input units and hidden units, thus having a relatively small number of parameters. Each dimension of $\\theta$ requires two evaluations of the cost function and this can be expensive. In the function `checkNNGradients`, our code creates a small random model and dataset which is used with `computeNumericalGradient` for gradient checking. Furthermore, after we are confident that our gradient computations are correct, we should turn off gradient checking before running our learning algorithm.\n",
    "\n",
    "2. Gradient checking works for any function where we are computing the cost and the gradient. Concretely, we can use the same `computeNumericalGradient` function to check if our gradient implementations for the other cases are correct too, e.g. logistic regression’s cost function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numerical Gradient:   0.006907, Analytical Gradient   0.006907\n",
      "Numerical Gradient:  -0.000111, Analytical Gradient  -0.000111\n",
      "Numerical Gradient:   0.000144, Analytical Gradient   0.000144\n",
      "Numerical Gradient:   0.000266, Analytical Gradient   0.000266\n",
      "Numerical Gradient:   0.012125, Analytical Gradient   0.012125\n",
      "Numerical Gradient:  -0.000043, Analytical Gradient  -0.000043\n",
      "Numerical Gradient:   0.000152, Analytical Gradient   0.000152\n",
      "Numerical Gradient:   0.000207, Analytical Gradient   0.000207\n",
      "Numerical Gradient:   0.006198, Analytical Gradient   0.006198\n",
      "Numerical Gradient:   0.000064, Analytical Gradient   0.000064\n",
      "Numerical Gradient:   0.000020, Analytical Gradient   0.000020\n",
      "Numerical Gradient:  -0.000042, Analytical Gradient  -0.000042\n",
      "Numerical Gradient:  -0.005424, Analytical Gradient  -0.005424\n",
      "Numerical Gradient:   0.000113, Analytical Gradient   0.000113\n",
      "Numerical Gradient:  -0.000130, Analytical Gradient  -0.000130\n",
      "Numerical Gradient:  -0.000253, Analytical Gradient  -0.000253\n",
      "Numerical Gradient:  -0.012076, Analytical Gradient  -0.012076\n",
      "Numerical Gradient:   0.000057, Analytical Gradient   0.000057\n",
      "Numerical Gradient:  -0.000161, Analytical Gradient  -0.000161\n",
      "Numerical Gradient:  -0.000231, Analytical Gradient  -0.000231\n",
      "Numerical Gradient:   0.102286, Analytical Gradient   0.102286\n",
      "Numerical Gradient:   0.050959, Analytical Gradient   0.050959\n",
      "Numerical Gradient:   0.049651, Analytical Gradient   0.049651\n",
      "Numerical Gradient:   0.053277, Analytical Gradient   0.053277\n",
      "Numerical Gradient:   0.049845, Analytical Gradient   0.049845\n",
      "Numerical Gradient:   0.050705, Analytical Gradient   0.050705\n",
      "Numerical Gradient:   0.099593, Analytical Gradient   0.099593\n",
      "Numerical Gradient:   0.050232, Analytical Gradient   0.050232\n",
      "Numerical Gradient:   0.047839, Analytical Gradient   0.047839\n",
      "Numerical Gradient:   0.051921, Analytical Gradient   0.051921\n",
      "Numerical Gradient:   0.048977, Analytical Gradient   0.048977\n",
      "Numerical Gradient:   0.048743, Analytical Gradient   0.048743\n",
      "Numerical Gradient:   0.296932, Analytical Gradient   0.296932\n",
      "Numerical Gradient:   0.148333, Analytical Gradient   0.148333\n",
      "Numerical Gradient:   0.142656, Analytical Gradient   0.142656\n",
      "Numerical Gradient:   0.156193, Analytical Gradient   0.156193\n",
      "Numerical Gradient:   0.144171, Analytical Gradient   0.144171\n",
      "Numerical Gradient:   0.146350, Analytical Gradient   0.146350\n",
      "\n",
      "The above two columns should be very similar.\n",
      "If the backpropagation implementation is correct, then \n",
      "the relative difference will be small (less than 1e-9). \n",
      "\n",
      "Relative Difference: 1.8750864672E-11\n"
     ]
    }
   ],
   "source": [
    "def debugInitializeWeights(fan_out, fan_in):\n",
    "    \"\"\"\n",
    "    Initialize the weights of a layer with fan_in\n",
    "    incoming connections and fan_out outgoing\n",
    "    connections using a fix set of values. This will\n",
    "    help us in debugging.\n",
    "    Note that W should be set to a matrix of size\n",
    "    (1 + fan_in, fan_out) as the first row of W\n",
    "    handles the \"bias\" terms.\n",
    "    Args:\n",
    "        fan_in: int\n",
    "        fan_out: int\n",
    "    Returns:\n",
    "        W: array (fan_out, 1 + fan_in)\n",
    "    \"\"\"\n",
    "    # Initialize the W matrix.\n",
    "    W = np.zeros((fan_out, 1 + fan_in))\n",
    "    \n",
    "    # The \"sin\" function ensures that W is always of\n",
    "    # the same values and will be useful for debugging.\n",
    "    W = np.reshape(np.sin(range(W.size)), W.shape) / 10\n",
    "    \n",
    "    return W\n",
    "\n",
    "def computeNumericalGradient(J, theta):\n",
    "    \"\"\"\n",
    "    Computes the numerical gradient of the function J\n",
    "    around theta using \"finite differences\" and gives\n",
    "    a numerical estimate of the gradient.\n",
    "    Notes: The following code implements numerical\n",
    "           gradient checking, and returns the numerical\n",
    "           gradient. It sets numgrad(i) to (a numerical \n",
    "           approximation of) the partial derivative of J\n",
    "           with respect to the i-th input argument,\n",
    "           evaluated at theta. (i.e., numgrad(i) should\n",
    "           be the (approximately) the partial derivative\n",
    "           of J with respect to theta(i).)\n",
    "    Args:\n",
    "        J: function\n",
    "        theta: vector array\n",
    "    Returns:\n",
    "        numgrad: vector array\n",
    "    \"\"\"\n",
    "    # Initialize parameters.\n",
    "    numgrad = np.zeros(theta.shape)\n",
    "    perturb = np.zeros(theta.shape)\n",
    "    e = 1e-4\n",
    "\n",
    "    for p in range(theta.size):\n",
    "        # Set the perturbation vector.\n",
    "        perturb.reshape(perturb.size)[p] = e\n",
    "        loss1, _ = J(theta - perturb)\n",
    "        loss2, _ = J(theta + perturb)\n",
    "        # Compute the Numerical Gradient.\n",
    "        numgrad.reshape(numgrad.size)[p] = (loss2 - loss1) / (2 * e)\n",
    "        perturb.reshape(perturb.size)[p] = 0\n",
    "        \n",
    "    return numgrad\n",
    "\n",
    "def checkNNGradients(lambda_coef):\n",
    "    \"\"\"\n",
    "    Creates a small neural network to check the\n",
    "    backpropagation gradients. It will output the\n",
    "    analytical gradients produced by the backprop\n",
    "    code and the numerical gradients (computed\n",
    "    using computeNumericalGradient). These two\n",
    "    gradient computations should result in very\n",
    "    similar values.\n",
    "    Args:\n",
    "        lambda_coef: int\n",
    "    \"\"\"\n",
    "    # Initialize a small network.\n",
    "    input_layer_size = 3\n",
    "    hidden_layer_size = 5\n",
    "    num_labels = 3\n",
    "    m = 5\n",
    "\n",
    "    # Generate some 'random' test data.\n",
    "    Theta1 = debugInitializeWeights(hidden_layer_size, input_layer_size)\n",
    "    Theta2 = debugInitializeWeights(num_labels, hidden_layer_size)\n",
    "    \n",
    "    # Reuse debugInitializeWeights to generate X.\n",
    "    X  = debugInitializeWeights(m, input_layer_size - 1)\n",
    "    # np.mod returns element-wise remainder of division.\n",
    "    y  = 1 + np.mod(range(m), num_labels).T\n",
    "\n",
    "    # Unroll parameters.\n",
    "    Thetas = [Theta1, Theta2]\n",
    "    unrolled_Thetas = [Thetas[i].ravel() for i,_ in enumerate(Thetas)]\n",
    "    nn_params = np.concatenate(unrolled_Thetas)\n",
    "\n",
    "    # Create short hand for cost function.\n",
    "    def costFunc(p):\n",
    "        return nnCostFunction(p, input_layer_size, hidden_layer_size,\n",
    "                              num_labels, X, y, lambda_coef)\n",
    "\n",
    "    _, grad = costFunc(nn_params)\n",
    "    numgrad = computeNumericalGradient(costFunc, nn_params)\n",
    "\n",
    "    # Visually examine the two gradient computations.\n",
    "    # The two columns should be very similar. \n",
    "    for numerical, analytical in zip(numgrad, grad):\n",
    "        print('Numerical Gradient: {0:10f}, Analytical Gradient {1:10f}'.format(numerical, analytical))\n",
    "    print('\\nThe above two columns should be very similar.')\n",
    "\n",
    "    # Evaluate the norm of the difference between two solutions.  \n",
    "    # If we have a correct implementation, and assuming we used e = 0.0001 \n",
    "    # in computeNumericalGradient, then diff below should be less than 1e-9.\n",
    "    diff = np.linalg.norm(numgrad-grad) / np.linalg.norm(numgrad+grad)\n",
    "\n",
    "    print('If the backpropagation implementation is correct, then \\n' \\\n",
    "             'the relative difference will be small (less than 1e-9). \\n' \\\n",
    "             '\\nRelative Difference: {:.10E}'.format(diff))\n",
    "\n",
    "checkNNGradients(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Learning Parameters\n",
    "\n",
    "After we have successfully implemented the neural network cost function and gradient computation, the next step is to use an optimization function such as `minimize` to learn a good set parameters.\n",
    "\n",
    "After the training completes, we can proceed to report the training accuracy of our classifier by computing the percentage of examples it got correct. If our implementation is correct, we should see a reported training accuracy of about $95\\%$ (this may vary by about $1%$ due to the random initialization). It is possible to get higher training accuracies by training the neural network for more iterations. We can try training the neural network for more iterations (e.g., set `MaxIter` to `400`) and also vary the regularization parameter $\\lambda$. With the right learning settings, it is possible to get the neural network to perfectly fit the training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 0.490673\n",
      "         Iterations: 40\n",
      "         Function evaluations: 84\n",
      "         Gradient evaluations: 84\n"
     ]
    }
   ],
   "source": [
    "from scipy.optimize import minimize\n",
    "\n",
    "# Create a cost function optimizer.\n",
    "def optimizeCost(maxiter, lambda_coef):\n",
    "    \"\"\"\n",
    "    Optimizes the Cost Function.\n",
    "    Args:\n",
    "        maxiter: int\n",
    "        lambda_coef: float\n",
    "    Returns:\n",
    "        thetas: vector array\n",
    "    \"\"\"\n",
    "    ##########################################\n",
    "    ########## Initialize nn_params ##########\n",
    "    ##########################################\n",
    "    Theta1 = randInitializeWeights(input_layer_size, hidden_layer_size)\n",
    "    Theta2 = randInitializeWeights(hidden_layer_size, num_labels)\n",
    "    # Unroll parameters.\n",
    "    Thetas = [Theta1, Theta2]\n",
    "    unrolled_Thetas = [Thetas[i].ravel() for i,_ in enumerate(Thetas)]\n",
    "    nn_params = np.concatenate(unrolled_Thetas)\n",
    "    \n",
    "    ##########################################\n",
    "    ##########       Optimize       ##########\n",
    "    ##########################################\n",
    "    # Optimize the cost function.\n",
    "    results = minimize(fun=nnCostFunction,\n",
    "                       x0=nn_params,\n",
    "                       args=(input_layer_size,\n",
    "                             hidden_layer_size,\n",
    "                             num_labels,\n",
    "                             X, y, lambda_coef),\n",
    "                       method='CG',\n",
    "                       jac=True,\n",
    "                       options={'maxiter':maxiter, 'disp': True})\n",
    "    thetas = results.x\n",
    "\n",
    "    return thetas\n",
    "\n",
    "# Train the neural network for 40 iterations.\n",
    "# Set regularization parameter to 1.\n",
    "nn_params = optimizeCost(40, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Training Set Accuracy: 94.80%\n"
     ]
    }
   ],
   "source": [
    "# Create the prediction function\n",
    "def predict(Theta1, Theta2, X):\n",
    "    \"\"\"\n",
    "    Predicts the label of an input X\n",
    "    given the trained weights of a\n",
    "    neural network (Theta1, Theta2).\n",
    "    Args:\n",
    "        Theta1: array (hidden_layer_size, 1 + input_layer_size)\n",
    "        Theta2: array (num_labels, 1 + hidden_layer_size)\n",
    "        X: array (m, input_layer_size)\n",
    "    Returns:\n",
    "        p_argmax: vector array (m,)\n",
    "    \"\"\"\n",
    "    # Get some useful values.\n",
    "    m = len(X)\n",
    "    num_labels = len(Theta2)\n",
    "    \n",
    "    # Initialize prediction p.\n",
    "    p = np.zeros((m, 1))\n",
    "    \n",
    "    #Insert a 1's column to X.\n",
    "    X = np.insert(X, 0, 1, axis=1)\n",
    "    a2 = sigmoid(np.dot(X, Theta1.T))\n",
    "    #Insert a 1's column to a2.\n",
    "    a2 = np.insert(a2, 0, 1, axis=1)\n",
    "    h_theta = sigmoid(np.dot(a2, Theta2.T))\n",
    "    p_argmax = np.argmax(h_theta, axis=1)\n",
    "    # Add 1 to fix for Python's zero indexed array.\n",
    "    p_argmax = p_argmax + 1\n",
    "    \n",
    "    return p_argmax\n",
    "\n",
    "# Create a function to output the accuracy score.\n",
    "def outputAccuracy(nn_params):\n",
    "    \"\"\"\n",
    "    Outputs the accuracy of the\n",
    "    trained neural network.\n",
    "    Args:\n",
    "        nn_params: vector array\n",
    "    \"\"\"\n",
    "    # After the training of the neural network, reshape\n",
    "    # nn_params back into the parameters Theta1 and Theta2,\n",
    "    # the weight matrices for our 2 layer neural network.\n",
    "    Theta1 = np.reshape(nn_params[:hidden_layer_size * (input_layer_size + 1)],\n",
    "                        (hidden_layer_size, input_layer_size + 1))\n",
    "    Theta2 = np.reshape(nn_params[hidden_layer_size * (input_layer_size + 1):],\n",
    "                        (num_labels, hidden_layer_size + 1))\n",
    "\n",
    "    predictions = predict(Theta1, Theta2, X)\n",
    "    print('\\nTraining Set Accuracy: {0:.2f}%'.format(100 * np.mean(predictions == y.reshape(-1))))\n",
    "\n",
    "outputAccuracy(nn_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing the Hidden Layer\n",
    "\n",
    "One way to understand what our neural network is learning is to visualize what the representations captured by the hidden units. Informally, given a particular hidden unit, one way to visualize what it computes is to find an input $x$ that will cause it to activate (that is, to have an activation value $(\\alpha^{(l)})$ close to $1$). For the neural network we trained, notice that the $i^{th}$ row of $\\Theta^{(1)}$ is a 401-dimensional vector that represents the parameter for the $i^{th}$ hidden unit. If we discard the bias term, we get a 400-dimensional vector that represents the weights from each input pixel to the hidden unit.\n",
    "\n",
    "Thus, one way to visualize the “representation” captured by the hidden unit is to reshape this 400-dimensional vector into a $20x20$ image and display it. This can be done by using the `displayData` function, which will show an image with 25 units, each corresponding to one hidden unit in the network.\n",
    "\n",
    "In the trained network, it can be found that the hidden units corresponds roughly to detectors that look for strokes and other patterns in the input."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWMAAAFbCAYAAAAA6+8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x181fd46fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "displayData(Theta1[:, 1:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Play\n",
    "\n",
    "We can try out different learning settings for the neural network to see how the performance of the neural network varies with the regularization parameter $\\lambda$ and number of training steps (the `MaxIter` option when using `minimize`).\n",
    "\n",
    "Neural networks are very powerful models that can form highly complex decision boundaries. Without regularization, it is possible for a neural network to “overfit” a training set so that it obtains close to $100\\%$ accuracy on the training set but does not as well on new examples that it has not seen before. The regularization $\\lambda$ can be set to a smaller value and the `MaxIter` parameter to a higher number of iterations to see this.\n",
    "\n",
    "The changes in the visualizations of the hidden units can also be depicted when the learning parameters $\\lambda$ and `MaxIter` are changed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------\n",
      "\n",
      "MaxIter = 10 and lambda_coef = 0.5\n",
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 1.256776\n",
      "         Iterations: 10\n",
      "         Function evaluations: 21\n",
      "         Gradient evaluations: 21\n",
      "\n",
      "Training Set Accuracy: 82.74%\n",
      "------------------------------------------------------------\n",
      "\n",
      "MaxIter = 10 and lambda_coef = 100\n",
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 2.702304\n",
      "         Iterations: 10\n",
      "         Function evaluations: 22\n",
      "         Gradient evaluations: 22\n",
      "\n",
      "Training Set Accuracy: 73.10%\n",
      "------------------------------------------------------------\n",
      "\n",
      "MaxIter = 10 and lambda_coef = 1000\n",
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 3.292116\n",
      "         Iterations: 10\n",
      "         Function evaluations: 20\n",
      "         Gradient evaluations: 20\n",
      "\n",
      "Training Set Accuracy: 10.00%\n",
      "------------------------------------------------------------\n",
      "\n",
      "MaxIter = 30 and lambda_coef = 1\n",
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 0.674454\n",
      "         Iterations: 30\n",
      "         Function evaluations: 63\n",
      "         Gradient evaluations: 63\n",
      "\n",
      "Training Set Accuracy: 92.90%\n",
      "------------------------------------------------------------\n",
      "\n",
      "MaxIter = 200 and lambda_coef = 0\n",
      "Warning: Maximum number of iterations has been exceeded.\n",
      "         Current function value: 0.007894\n",
      "         Iterations: 200\n",
      "         Function evaluations: 581\n",
      "         Gradient evaluations: 581\n",
      "\n",
      "Training Set Accuracy: 100.00%\n"
     ]
    }
   ],
   "source": [
    "# Create different cases of the learning parameters, lambda_coef and MaxIter.\n",
    "cases = {\"MaxIter = 10 and lambda_coef = 0.5\": (10, 0.5),\n",
    "         \"MaxIter = 10 and lambda_coef = 100\": (10, 100),\n",
    "         \"MaxIter = 10 and lambda_coef = 1000\": (10, 1000),\n",
    "         \"MaxIter = 30 and lambda_coef = 1\": (30, 1),\n",
    "         \"MaxIter = 200 and lambda_coef = 0\": (200, 0)}\n",
    "\n",
    "for k, v in cases.items():\n",
    "    print(60 * \"-\")\n",
    "    print()\n",
    "    print(k)\n",
    "    nn_params = optimizeCost(v[0], v[1])\n",
    "    outputAccuracy(nn_params)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}