Traffic_Sign_Classifier-final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Self-Driving Car Engineer Nanodegree\n",
    "\n",
    "## Deep Learning\n",
    "\n",
    "## Project: Build a Traffic Sign Recognition Classifier\n",
    "\n",
    "In this notebook, a template is provided for you to implement your functionality in stages, which is required to successfully complete this project. If additional code is required that cannot be included in the notebook, be sure that the Python code is successfully imported and included in your submission if necessary. \n",
    "\n",
    "> **Note**: Once you have completed all of the code implementations, you need to finalize your work by exporting the iPython Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to  \\n\",\n",
    "    \"**File -> Download as -> HTML (.html)**. Include the finished document along with this notebook as your submission. \n",
    "\n",
    "In addition to implementing code, there is a writeup to complete. The writeup should be completed in a separate file, which can be either a markdown file or a pdf document. There is a [write up template](https://github.com/udacity/CarND-Traffic-Sign-Classifier-Project/blob/master/writeup_template.md) that can be used to guide the writing process. Completing the code template and writeup template will cover all of the [rubric points](https://review.udacity.com/#!/rubrics/481/view) for this project.\n",
    "\n",
    "The [rubric](https://review.udacity.com/#!/rubrics/481/view) contains \"Stand Out Suggestions\" for enhancing the project beyond the minimum requirements. The stand out suggestions are optional. If you decide to pursue the \"stand out suggestions\", you can include the code in this Ipython notebook and also discuss the results in the writeup file.\n",
    "\n",
    "\n",
    ">**Note:** Code and Markdown cells can be executed using the **Shift + Enter** keyboard shortcut. In addition, Markdown cells can be edited by typically double-clicking the cell to enter edit mode."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Step 0: Load The Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# import statements\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import skimage.color as color\n",
    "import pickle\n",
    "import math\n",
    "from sklearn.utils import shuffle\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Load pickled data\n",
    "\n",
    "training_file = 'train.p'\n",
    "validation_file= 'valid.p'\n",
    "testing_file = 'test.p'\n",
    "\n",
    "with open(training_file, mode='rb') as f:\n",
    "    train = pickle.load(f)\n",
    "with open(validation_file, mode='rb') as f:\n",
    "    valid = pickle.load(f)\n",
    "with open(testing_file, mode='rb') as f:\n",
    "    test = pickle.load(f)\n",
    "    \n",
    "X_train, y_train = train['features'], train['labels']\n",
    "X_valid, y_valid = valid['features'], valid['labels']\n",
    "X_test, y_test = test['features'], test['labels']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Image Shape: (32, 32, 3)\n",
      "\n",
      "Training Set:   34799 samples\n",
      "Validation Set: 4410 samples\n",
      "Test Set:       12630 samples\n"
     ]
    }
   ],
   "source": [
    "print()\n",
    "print(\"Image Shape: {}\".format(X_train[0].shape))\n",
    "print()\n",
    "print(\"Training Set:   {} samples\".format(len(X_train)))\n",
    "print(\"Validation Set: {} samples\".format(len(X_valid)))\n",
    "print(\"Test Set:       {} samples\".format(len(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "helper functions defined\n"
     ]
    }
   ],
   "source": [
    "# helper functions\n",
    "def print_random_images(first_set, labels_set, second_set = None, count = 4):\n",
    "    \"\"\"prints random images from the given first and second set, count is how many images to take\n",
    "    \"\"\"\n",
    "    images = []\n",
    "    labels = []\n",
    "    if second_set is None:\n",
    "        rows = count//4\n",
    "        cols = count//2\n",
    "    else:\n",
    "        rows = count//2\n",
    "        cols = count//2\n",
    "        \n",
    "    for i in range(count):\n",
    "        index = random.randint(0, len(first_set) - 1)\n",
    "        images.append(first_set[index])\n",
    "        labels.append(labels_set[index])\n",
    "        if second_set is not None:\n",
    "            images.append(second_set[index])\n",
    "            labels.append(labels_set[index])\n",
    "           \n",
    "    print_images(images, labels, row=rows, col=cols)\n",
    "    \n",
    "def print_images(data, labels, row = 1, col = 1, cmap = 'gray'):\n",
    "    \"\"\"print images from given data array\n",
    "    \"\"\"\n",
    "    row = np.max([math.ceil(row), 1])\n",
    "    col = np.max([math.ceil(col), 1])\n",
    "    if row == 1 and col == 1: \n",
    "        plt.figure(figsize=(1,1))\n",
    "        plt.imshow(data[0], cmap=cmap)\n",
    "    else:\n",
    "        row_col = row*col\n",
    "        fig, axs = plt.subplots(row, col, figsize=(row_col, row_col))\n",
    "        fig.subplots_adjust(hspace = .2, wspace=.1)\n",
    "        axs = axs.ravel()\n",
    "        total = np.min([len(data), row_col])\n",
    "        for x in range(total):\n",
    "#             print('x: ', x, 'data[x].shape', data[x].shape)\n",
    "            axs[x].axis('off')\n",
    "            if labels is not None:\n",
    "                axs[x].set_title(labels[x])\n",
    "            if cmap is None:\n",
    "                axs[x].imshow(data[x].squeeze())  \n",
    "            else:\n",
    "                axs[x].imshow(data[x].squeeze(), cmap=cmap)  \n",
    "\n",
    "def normalize(data):\n",
    "    \"\"\"Normalize the given data\"\"\"\n",
    "    return (data - np.min(data))/ (np.max(data) - np.min(data))\n",
    "\n",
    "def convert_to_grayscale(data):\n",
    "    \"\"\"Convert the given images data to grayscale, using the average of the RGB channel. \n",
    "    Returns one dimension channel instead of 3\"\"\"\n",
    "    return np.sum(data/3, axis=3, keepdims=True)\n",
    "\n",
    "            \n",
    "print('helper functions defined')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Step 1: Dataset Summary & Exploration\n",
    "\n",
    "The pickled data is a dictionary with 4 key/value pairs:\n",
    "\n",
    "- `'features'` is a 4D array containing raw pixel data of the traffic sign images, (num examples, width, height, channels).\n",
    "- `'labels'` is a 1D array containing the label/class id of the traffic sign. The file `signnames.csv` contains id -> name mappings for each id.\n",
    "- `'sizes'` is a list containing tuples, (width, height) representing the original width and height the image.\n",
    "- `'coords'` is a list containing tuples, (x1, y1, x2, y2) representing coordinates of a bounding box around the sign in the image. **THESE COORDINATES ASSUME THE ORIGINAL IMAGE. THE PICKLED DATA CONTAINS RESIZED VERSIONS (32 by 32) OF THESE IMAGES**\n",
    "\n",
    "Complete the basic data summary below. Use python, numpy and/or pandas methods to calculate the data summary rather than hard coding the results. For example, the [pandas shape method](http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.shape.html) might be useful for calculating some of the summary results. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Provide a Basic Summary of the Data Set Using Python, Numpy and/or Pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training examples = 34799\n",
      "Number of testing examples = 12630\n",
      "Image data shape = (32, 32, 3)\n",
      "Number of classes = 43\n"
     ]
    }
   ],
   "source": [
    "### Replace each question mark with the appropriate value. \n",
    "### Use python, pandas or numpy methods rather than hard coding the results\n",
    "\n",
    "# TODO: Number of training examples\n",
    "n_train = len(X_train)\n",
    "\n",
    "# TODO: Number of validation examples\n",
    "n_validation = len(X_valid)\n",
    "\n",
    "# TODO: Number of testing examples.\n",
    "n_test = len(X_test)\n",
    "\n",
    "# TODO: What's the shape of an traffic sign image?\n",
    "image_shape = X_train[0].shape\n",
    "\n",
    "# TODO: How many unique classes/labels there are in the dataset.\n",
    "n_classes = len(np.unique(y_train))\n",
    "\n",
    "print(\"Number of training examples =\", n_train)\n",
    "print(\"Number of testing examples =\", n_test)\n",
    "print(\"Image data shape =\", image_shape)\n",
    "print(\"Number of classes =\", n_classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Include an exploratory visualization of the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize the German Traffic Signs Dataset using the pickled file(s). This is open ended, suggestions include: plotting traffic sign images, plotting the count of each sign, etc. \n",
    "\n",
    "The [Matplotlib](http://matplotlib.org/) [examples](http://matplotlib.org/examples/index.html) and [gallery](http://matplotlib.org/gallery.html) pages are a great resource for doing visualizations in Python.\n",
    "\n",
    "**NOTE:** It's recommended you start with something simple first. If you wish to do more, come back to it after you've completed the rest of the sections. It can be interesting to look at the distribution of classes in the training, validation and test set. Is the distribution the same? Are there more examples of some classes than others?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAGECAYAAADnWlOjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27f2d53f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_random_images(X_train, y_train, count=8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----\n",
    "\n",
    "## Step 2: Design and Test a Model Architecture\n",
    "\n",
    "Design and implement a deep learning model that learns to recognize traffic signs. Train and test your model on the [German Traffic Sign Dataset](http://benchmark.ini.rub.de/?section=gtsrb&subsection=dataset).\n",
    "\n",
    "The LeNet-5 implementation shown in the [classroom](https://classroom.udacity.com/nanodegrees/nd013/parts/fbf77062-5703-404e-b60c-95b78b2f3f9e/modules/6df7ae49-c61c-4bb2-a23e-6527e69209ec/lessons/601ae704-1035-4287-8b11-e2c2716217ad/concepts/d4aca031-508f-4e0b-b493-e7b706120f81) at the end of the CNN lesson is a solid starting point. You'll have to change the number of classes and possibly the preprocessing, but aside from that it's plug and play! \n",
    "\n",
    "With the LeNet-5 solution from the lecture, you should expect a validation set accuracy of about 0.89. To meet specifications, the validation set accuracy will need to be at least 0.93. It is possible to get an even higher accuracy, but 0.93 is the minimum for a successful project submission. \n",
    "\n",
    "There are various aspects to consider when thinking about this problem:\n",
    "\n",
    "- Neural network architecture (is the network over or underfitting?)\n",
    "- Play around preprocessing techniques (normalization, rgb to grayscale, etc)\n",
    "- Number of examples per label (some have more than others).\n",
    "- Generate fake data.\n",
    "\n",
    "Here is an example of a [published baseline model on this problem](http://yann.lecun.com/exdb/publis/pdf/sermanet-ijcnn-11.pdf). It's not required to be familiar with the approach used in the paper but, it's good practice to try to read papers like these."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pre-process the Data Set (normalization, grayscale, etc.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Minimally, the image data should be normalized so that the data has mean zero and equal variance. For image data, `(pixel - 128)/ 128` is a quick way to approximately normalize the data and can be used in this project. \n",
    "\n",
    "Other pre-processing steps are optional. You can try different techniques to see if it improves performance. \n",
    "\n",
    "Use the code cell (or multiple code cells, if necessary) to implement the first step of your project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Image data shape = (32, 32, 1)\n",
      "Valid Image data shape = (32, 32, 1)\n",
      "Test Image data shape = (32, 32, 1)\n",
      "done\n"
     ]
    }
   ],
   "source": [
    "### Preprocess the data here. It is required to normalize the data. Other preprocessing steps could include \n",
    "### converting to grayscale, etc.\n",
    "### Feel free to use as many code cells as needed.\n",
    "\n",
    "#convert to grayscale\n",
    "X_train_gray = convert_to_grayscale(X_train)\n",
    "X_valid_gray = convert_to_grayscale(X_valid)\n",
    "X_test_gray = convert_to_grayscale(X_test)\n",
    "\n",
    "print(\"Train Image data shape =\", X_train_gray[0].shape)\n",
    "print(\"Valid Image data shape =\", X_valid_gray[0].shape)\n",
    "print(\"Test Image data shape =\", X_test_gray[0].shape)\n",
    "print('done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Showing grayscaled images along with its color counterpart\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA48AAAOVCAYAAADJJNXyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDanvzNcaAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27f2949cc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Showing grayscaled images along with its color counterpart')\n",
    "print_random_images(X_train, y_train, X_train_gray, 8)\n",
    "\n",
    "X_train = X_train_gray\n",
    "X_valid = X_valid_gray\n",
    "X_test = X_test_gray"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean before normalization\n",
      "train:  82.677589037\n",
      "valid:  83.5564273756\n",
      "test:   82.1484603612\n",
      "\n",
      "mean after normalization\n",
      "train:  0.314367065134\n",
      "valid:  0.314225709503\n",
      "test:   0.311348447654\n"
     ]
    }
   ],
   "source": [
    "#normalize the data\n",
    "\n",
    "print('mean before normalization')\n",
    "print('train: ', np.mean(X_train))\n",
    "print('valid: ', np.mean(X_valid))\n",
    "print('test:  ', np.mean(X_test))\n",
    "\n",
    "X_train_norm = normalize(X_train)\n",
    "X_valid_norm = normalize(X_valid)\n",
    "X_test_norm = normalize(X_test)\n",
    "\n",
    "print('\\nmean after normalization')\n",
    "print('train: ', np.mean(X_train_norm))\n",
    "print('valid: ', np.mean(X_valid_norm))\n",
    "print('test:  ', np.mean(X_test_norm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Analyzing images before and after normalization. Normalization should not alter the images\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA48AAAOVCAYAAADJJNXyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAACdbB4BAADoZPMIAABAp/8PMTyfcbDsMCgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f28741f7a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Analyzing images before and after normalization. Normalization should not alter the images\n",
    "print('Analyzing images before and after normalization. Normalization should not alter the images')\n",
    "print_random_images(X_train, y_train, X_train_norm, 8)\n",
    "\n",
    "X_train = X_train_norm\n",
    "X_valid = X_valid_norm\n",
    "X_test = X_test_norm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Augmentation\n",
    "I did not choose to do Augmentation, since without the Augmentation I was getting good accuracy.   \n",
    "\n",
    "But augmentation should be used whenever the training data is less or the distribution of the data is not even. Like if some classes have higher number of training data while some has less. Using Augmentation the training data can be increased\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Architecture\n",
    "\n",
    "I found LeNet to be very simple for this project, hence I used 4 Convolutional Layers and 4 Fully Connected layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Channel count:  1\n"
     ]
    }
   ],
   "source": [
    "### Define your architecture here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import tensorflow as tf\n",
    "\n",
    "EPOCHS = 90\n",
    "BATCH_SIZE = 128\n",
    "FC1 = 320\n",
    "FC2 = 150\n",
    "FC3 = 80\n",
    "_, _, CHANNEL = X_train[0].shape\n",
    "filter1 = 6\n",
    "filter2 = 16\n",
    "filter3 = 26\n",
    "filter4 = 42\n",
    "# filter5 = 30\n",
    "print('Channel count: ', CHANNEL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# functions for architecture\n",
    "def create_weights(shape):\n",
    "    sigma = 0.1\n",
    "    mu = 0\n",
    "#     print('create weight: ', shape)\n",
    "    return tf.Variable(tf.truncated_normal(shape, mean = mu, stddev = sigma))\n",
    " \n",
    "def create_biases(size):\n",
    "    return tf.Variable(tf.constant(0.05, shape=[size]))\n",
    "\n",
    "def create_conv_layer(input,\n",
    "               input_channel,        \n",
    "               filter_count,\n",
    "               filter_size = 5,\n",
    "               drop_out = 1,\n",
    "               padding='VALID',\n",
    "               do_max_pool=True):  \n",
    "    \"\"\"Create the convolutional layer\"\"\"\n",
    "    strides = [1, 1, 1, 1]\n",
    "    mu = 0\n",
    "    sigma = 0.1\n",
    "\n",
    "    weights = create_weights(shape=[filter_size, filter_size, input_channel, filter_count])\n",
    "    biases = create_biases(filter_count)\n",
    "\n",
    "    ## Creating the convolutional layer\n",
    "    layer = tf.nn.conv2d(input,\n",
    "                     weights,\n",
    "                     strides=strides,\n",
    "                     padding=padding) + biases\n",
    "    if do_max_pool:\n",
    "        layer = tf.nn.max_pool(value=layer,\n",
    "                            ksize=[1, 2, 2, 1],\n",
    "                            strides=[1, 2, 2, 1],\n",
    "                            padding=padding)\n",
    "        print('Max Pool done')\n",
    "\n",
    "    layer = tf.nn.relu(layer)\n",
    "\n",
    "    print('Conv Layer output: ', layer.get_shape())\n",
    "    return layer\n",
    "\n",
    "def create_connected_layer(input,\n",
    "                            num_inputs,\n",
    "                            num_outputs,\n",
    "                            use_relu=True):\n",
    "    \"\"\"Create fully connected layer\"\"\"\n",
    "    weights = create_weights(shape=[num_inputs, num_outputs])\n",
    "    biases = create_biases(num_outputs)\n",
    "    layer = tf.add(tf.matmul(input, weights), biases)\n",
    "    if use_relu:\n",
    "        layer = tf.nn.relu(layer)\n",
    "        print('Relu done')\n",
    "    print('Fully connected Layer output: ', layer.get_shape())\n",
    "    return layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Convolution architecture\n",
    "\n",
    "from tensorflow.contrib.layers import flatten\n",
    "\n",
    "def get_logits(x):\n",
    "    \"\"\"Get the logits, modified the LeNet architecture\"\"\"\n",
    "    \n",
    "    #Layer 1: Convolutional. Input = 32x32xCHANNEL. Output = 30x30xfilter1.\n",
    "    c1 = create_conv_layer(x, CHANNEL, filter1, filter_size=3, do_max_pool=False)\n",
    "    \n",
    "    #Layer 2: Convolutional. Output = 14x14xfilter2.\n",
    "    c2 = create_conv_layer(c1, filter1, filter2, filter_size=3, do_max_pool=True)\n",
    "    \n",
    "    c2 = tf.nn.dropout(c2, drop_out_conv)\n",
    "    \n",
    "    #Layer 3: Convolutional. Output = 6x6xfilter3.\n",
    "    c3 = create_conv_layer(c2, filter2, filter3, padding='VALID', filter_size=3, do_max_pool=True)\n",
    "    \n",
    "    c3 = tf.nn.dropout(c3, drop_out_conv)\n",
    "\n",
    "    #Layer 4: Convolutional. Output = 6x6xfilter4.\n",
    "    c4 = create_conv_layer(c3, filter3, filter4, padding='SAME', filter_size=5, do_max_pool=False)\n",
    "    \n",
    "    \n",
    "    #Flatten. Input = 4x4xfilter3. Output = FC_input.\n",
    "    flatten = tf.contrib.layers.flatten(c4)\n",
    "    print('flatten shape: ', flatten.get_shape())\n",
    "    _,flatten_shape = flatten.get_shape()\n",
    "    \n",
    "    #Layer 5: Fully Connected. Input = FC_input. Output = FC1.\n",
    "    fc1 = create_connected_layer(flatten, int(flatten_shape), FC1)\n",
    "    \n",
    "    fc1 = tf.nn.dropout(fc1, drop_out_fc)\n",
    "    \n",
    "    #Layer 6: Fully Connected. Input = FC1. Output = FC2.\n",
    "    fc2 = create_connected_layer(fc1, FC1, FC2)\n",
    "    \n",
    "    fc2 = tf.nn.dropout(fc2, drop_out_fc)\n",
    "    \n",
    "    #Layer 7: Fully Connected. Input = FC2. Output = FC3.\n",
    "    fc3 = create_connected_layer(fc2, FC2, FC3)\n",
    "    \n",
    "    fc3 = tf.nn.dropout(fc3, drop_out_fc)\n",
    "    \n",
    "    #Layer 8: Fully Connected. Input = FC3. Output = 43.\n",
    "    logits = create_connected_layer(fc3, FC3, n_classes, False)\n",
    "    \n",
    "    return logits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x = tf.placeholder(tf.float32, (None, 32, 32, CHANNEL))\n",
    "y = tf.placeholder(tf.int32, (None))\n",
    "drop_out_conv = tf.placeholder(tf.float32)\n",
    "drop_out_fc = tf.placeholder(tf.float32)\n",
    "one_hot_y = tf.one_hot(y, n_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Conv Layer output:  (?, 30, 30, 6)\n",
      "Max Pool done\n",
      "Conv Layer output:  (?, 14, 14, 16)\n",
      "Max Pool done\n",
      "Conv Layer output:  (?, 6, 6, 26)\n",
      "Conv Layer output:  (?, 6, 6, 42)\n",
      "flatten shape:  (?, 1512)\n",
      "Relu done\n",
      "Fully connected Layer output:  (?, 320)\n",
      "Relu done\n",
      "Fully connected Layer output:  (?, 150)\n",
      "Relu done\n",
      "Fully connected Layer output:  (?, 80)\n",
      "Fully connected Layer output:  (?, 43)\n"
     ]
    }
   ],
   "source": [
    "rate = 0.001\n",
    "\n",
    "logits = get_logits(x)\n",
    "cross_entropy = tf.nn.softmax_cross_entropy_with_logits(labels=one_hot_y, logits=logits)\n",
    "loss_operation = tf.reduce_mean(cross_entropy)\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate = rate)\n",
    "training_operation = optimizer.minimize(loss_operation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train, Validate and Test the Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A validation set can be used to assess how well the model is performing. A low accuracy on the training and validation\n",
    "sets imply underfitting. A high accuracy on the training set but low accuracy on the validation set implies overfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(one_hot_y, 1))\n",
    "accuracy_operation = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))\n",
    "\n",
    "#returns total loss and total accuracy\n",
    "\n",
    "def evaluate(X_data, y_data):\n",
    "    num_examples = len(X_data)\n",
    "    total_accuracy = 0\n",
    "    total_loss = 0\n",
    "    sess = tf.get_default_session()\n",
    "    for offset in range(0, num_examples, BATCH_SIZE):\n",
    "        batch_x, batch_y = X_data[offset:offset+BATCH_SIZE], y_data[offset:offset+BATCH_SIZE]\n",
    "        loss, accuracy = sess.run([loss_operation, accuracy_operation], \\\n",
    "                                  feed_dict={x: batch_x, y: batch_y, drop_out_fc:1., drop_out_conv:1.})\n",
    "        total_accuracy += (accuracy * len(batch_x))\n",
    "        total_loss += (loss * len(batch_x))\n",
    "    return total_loss/num_examples, total_accuracy / num_examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "\n",
      "EPOCH 1, T Loss: 3.407, V Loss: 3.487, T Acc: 0.094, V Acc: 0.082\n",
      "EPOCH 2, T Loss: 2.152, V Loss: 2.260, T Acc: 0.354, V Acc: 0.326\n",
      "EPOCH 3, T Loss: 1.097, V Loss: 1.188, T Acc: 0.632, V Acc: 0.589\n",
      "EPOCH 4, T Loss: 0.603, V Loss: 0.674, T Acc: 0.804, V Acc: 0.771\n",
      "EPOCH 5, T Loss: 0.402, V Loss: 0.457, T Acc: 0.879, V Acc: 0.857\n",
      "EPOCH 6, T Loss: 0.272, V Loss: 0.321, T Acc: 0.907, V Acc: 0.887\n",
      "EPOCH 7, T Loss: 0.201, V Loss: 0.270, T Acc: 0.939, V Acc: 0.905\n",
      "EPOCH 8, T Loss: 0.151, V Loss: 0.209, T Acc: 0.946, V Acc: 0.916\n",
      "EPOCH 9, T Loss: 0.121, V Loss: 0.175, T Acc: 0.961, V Acc: 0.939\n",
      "EPOCH 10, T Loss: 0.105, V Loss: 0.165, T Acc: 0.967, V Acc: 0.943\n",
      "EPOCH 11, T Loss: 0.087, V Loss: 0.162, T Acc: 0.965, V Acc: 0.939\n",
      "EPOCH 12, T Loss: 0.076, V Loss: 0.139, T Acc: 0.968, V Acc: 0.947\n",
      "EPOCH 13, T Loss: 0.072, V Loss: 0.147, T Acc: 0.972, V Acc: 0.948\n",
      "EPOCH 14, T Loss: 0.063, V Loss: 0.121, T Acc: 0.975, V Acc: 0.955\n",
      "EPOCH 15, T Loss: 0.059, V Loss: 0.128, T Acc: 0.980, V Acc: 0.953\n",
      "EPOCH 16, T Loss: 0.050, V Loss: 0.121, T Acc: 0.979, V Acc: 0.958\n",
      "EPOCH 17, T Loss: 0.046, V Loss: 0.114, T Acc: 0.979, V Acc: 0.960\n",
      "EPOCH 18, T Loss: 0.040, V Loss: 0.127, T Acc: 0.981, V Acc: 0.957\n",
      "EPOCH 19, T Loss: 0.039, V Loss: 0.111, T Acc: 0.981, V Acc: 0.960\n",
      "EPOCH 20, T Loss: 0.039, V Loss: 0.139, T Acc: 0.980, V Acc: 0.953\n",
      "EPOCH 21, T Loss: 0.032, V Loss: 0.115, T Acc: 0.985, V Acc: 0.958\n",
      "EPOCH 22, T Loss: 0.030, V Loss: 0.120, T Acc: 0.983, V Acc: 0.958\n",
      "EPOCH 23, T Loss: 0.025, V Loss: 0.086, T Acc: 0.992, V Acc: 0.971\n",
      "EPOCH 24, T Loss: 0.024, V Loss: 0.100, T Acc: 0.993, V Acc: 0.965\n",
      "EPOCH 25, T Loss: 0.019, V Loss: 0.084, T Acc: 0.996, V Acc: 0.975\n",
      "EPOCH 26, T Loss: 0.018, V Loss: 0.120, T Acc: 0.995, V Acc: 0.967\n",
      "EPOCH 27, T Loss: 0.014, V Loss: 0.079, T Acc: 0.997, V Acc: 0.978\n",
      "EPOCH 28, T Loss: 0.011, V Loss: 0.116, T Acc: 0.998, V Acc: 0.970\n",
      "EPOCH 29, T Loss: 0.012, V Loss: 0.077, T Acc: 0.997, V Acc: 0.977\n",
      "EPOCH 30, T Loss: 0.013, V Loss: 0.115, T Acc: 0.996, V Acc: 0.966\n",
      "EPOCH 31, T Loss: 0.011, V Loss: 0.121, T Acc: 0.996, V Acc: 0.968\n",
      "EPOCH 32, T Loss: 0.008, V Loss: 0.121, T Acc: 0.997, V Acc: 0.972\n",
      "EPOCH 33, T Loss: 0.007, V Loss: 0.101, T Acc: 0.999, V Acc: 0.974\n",
      "EPOCH 34, T Loss: 0.007, V Loss: 0.090, T Acc: 0.998, V Acc: 0.976\n",
      "EPOCH 35, T Loss: 0.007, V Loss: 0.102, T Acc: 0.999, V Acc: 0.971\n",
      "EPOCH 36, T Loss: 0.005, V Loss: 0.069, T Acc: 0.999, V Acc: 0.980\n",
      "EPOCH 37, T Loss: 0.005, V Loss: 0.066, T Acc: 0.999, V Acc: 0.982\n",
      "EPOCH 38, T Loss: 0.005, V Loss: 0.059, T Acc: 0.999, V Acc: 0.982\n",
      "EPOCH 39, T Loss: 0.004, V Loss: 0.078, T Acc: 0.999, V Acc: 0.979\n",
      "EPOCH 40, T Loss: 0.006, V Loss: 0.105, T Acc: 0.998, V Acc: 0.971\n",
      "EPOCH 41, T Loss: 0.006, V Loss: 0.095, T Acc: 0.998, V Acc: 0.975\n",
      "EPOCH 42, T Loss: 0.003, V Loss: 0.075, T Acc: 0.999, V Acc: 0.978\n",
      "EPOCH 43, T Loss: 0.011, V Loss: 0.208, T Acc: 0.996, V Acc: 0.955\n",
      "EPOCH 44, T Loss: 0.004, V Loss: 0.077, T Acc: 0.999, V Acc: 0.977\n",
      "EPOCH 45, T Loss: 0.003, V Loss: 0.059, T Acc: 0.999, V Acc: 0.982\n",
      "EPOCH 46, T Loss: 0.005, V Loss: 0.072, T Acc: 0.998, V Acc: 0.980\n",
      "EPOCH 47, T Loss: 0.003, V Loss: 0.098, T Acc: 0.999, V Acc: 0.975\n",
      "EPOCH 48, T Loss: 0.005, V Loss: 0.123, T Acc: 0.999, V Acc: 0.971\n",
      "EPOCH 49, T Loss: 0.003, V Loss: 0.076, T Acc: 0.999, V Acc: 0.980\n",
      "EPOCH 50, T Loss: 0.002, V Loss: 0.068, T Acc: 0.999, V Acc: 0.981\n",
      "EPOCH 51, T Loss: 0.004, V Loss: 0.129, T Acc: 0.999, V Acc: 0.970\n",
      "EPOCH 52, T Loss: 0.002, V Loss: 0.090, T Acc: 1.000, V Acc: 0.980\n",
      "EPOCH 53, T Loss: 0.003, V Loss: 0.073, T Acc: 0.999, V Acc: 0.982\n",
      "EPOCH 54, T Loss: 0.002, V Loss: 0.096, T Acc: 0.999, V Acc: 0.977\n",
      "EPOCH 55, T Loss: 0.002, V Loss: 0.090, T Acc: 0.999, V Acc: 0.978\n",
      "EPOCH 56, T Loss: 0.003, V Loss: 0.114, T Acc: 0.999, V Acc: 0.973\n",
      "EPOCH 57, T Loss: 0.002, V Loss: 0.090, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 58, T Loss: 0.001, V Loss: 0.090, T Acc: 1.000, V Acc: 0.980\n",
      "EPOCH 59, T Loss: 0.002, V Loss: 0.089, T Acc: 0.999, V Acc: 0.976\n",
      "EPOCH 60, T Loss: 0.002, V Loss: 0.074, T Acc: 1.000, V Acc: 0.981\n",
      "EPOCH 61, T Loss: 0.001, V Loss: 0.105, T Acc: 1.000, V Acc: 0.976\n",
      "EPOCH 62, T Loss: 0.002, V Loss: 0.125, T Acc: 0.999, V Acc: 0.973\n",
      "EPOCH 63, T Loss: 0.002, V Loss: 0.129, T Acc: 0.999, V Acc: 0.972\n",
      "EPOCH 64, T Loss: 0.002, V Loss: 0.084, T Acc: 1.000, V Acc: 0.980\n",
      "EPOCH 65, T Loss: 0.001, V Loss: 0.084, T Acc: 1.000, V Acc: 0.979\n",
      "EPOCH 66, T Loss: 0.002, V Loss: 0.156, T Acc: 1.000, V Acc: 0.975\n",
      "EPOCH 67, T Loss: 0.002, V Loss: 0.105, T Acc: 0.999, V Acc: 0.976\n",
      "EPOCH 68, T Loss: 0.001, V Loss: 0.125, T Acc: 1.000, V Acc: 0.975\n",
      "EPOCH 69, T Loss: 0.001, V Loss: 0.098, T Acc: 1.000, V Acc: 0.979\n",
      "EPOCH 70, T Loss: 0.001, V Loss: 0.148, T Acc: 1.000, V Acc: 0.975\n",
      "EPOCH 71, T Loss: 0.002, V Loss: 0.142, T Acc: 1.000, V Acc: 0.974\n",
      "EPOCH 72, T Loss: 0.002, V Loss: 0.102, T Acc: 0.999, V Acc: 0.980\n",
      "EPOCH 73, T Loss: 0.001, V Loss: 0.083, T Acc: 1.000, V Acc: 0.982\n",
      "EPOCH 74, T Loss: 0.003, V Loss: 0.067, T Acc: 0.999, V Acc: 0.983\n",
      "EPOCH 75, T Loss: 0.002, V Loss: 0.084, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 76, T Loss: 0.001, V Loss: 0.080, T Acc: 1.000, V Acc: 0.983\n",
      "EPOCH 77, T Loss: 0.001, V Loss: 0.091, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 78, T Loss: 0.001, V Loss: 0.079, T Acc: 1.000, V Acc: 0.981\n",
      "EPOCH 79, T Loss: 0.001, V Loss: 0.108, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 80, T Loss: 0.001, V Loss: 0.116, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 81, T Loss: 0.002, V Loss: 0.156, T Acc: 1.000, V Acc: 0.970\n",
      "EPOCH 82, T Loss: 0.001, V Loss: 0.115, T Acc: 1.000, V Acc: 0.977\n",
      "EPOCH 83, T Loss: 0.001, V Loss: 0.122, T Acc: 1.000, V Acc: 0.975\n",
      "EPOCH 84, T Loss: 0.001, V Loss: 0.129, T Acc: 1.000, V Acc: 0.975\n",
      "EPOCH 85, T Loss: 0.001, V Loss: 0.121, T Acc: 1.000, V Acc: 0.976\n",
      "EPOCH 86, T Loss: 0.002, V Loss: 0.179, T Acc: 0.999, V Acc: 0.975\n",
      "EPOCH 87, T Loss: 0.001, V Loss: 0.092, T Acc: 1.000, V Acc: 0.983\n",
      "EPOCH 88, T Loss: 0.001, V Loss: 0.102, T Acc: 1.000, V Acc: 0.978\n",
      "EPOCH 89, T Loss: 0.001, V Loss: 0.099, T Acc: 1.000, V Acc: 0.980\n",
      "EPOCH 90, T Loss: 0.002, V Loss: 0.076, T Acc: 0.999, V Acc: 0.983\n",
      "Model saved with best validation accuracy of: 0.983\n"
     ]
    }
   ],
   "source": [
    "# #train model\n",
    "saver = tf.train.Saver()\n",
    "model_name = 'trafficSignModel'\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    \n",
    "    print(\"Training...\")\n",
    "    print()\n",
    "    valid_loss_all = []\n",
    "    train_loss_all = []\n",
    "    valid_acc_all = []\n",
    "    train_acc_all = []\n",
    "    best_valid_acc = 0\n",
    "    for i in range(EPOCHS):\n",
    "        X_train, y_train = shuffle(X_train, y_train)\n",
    "        for offset in range(0, n_train, BATCH_SIZE):\n",
    "            end = offset + BATCH_SIZE\n",
    "            batch_x, batch_y = X_train[offset:end], y_train[offset:end]\n",
    "            sess.run(training_operation, feed_dict={x: batch_x, y: batch_y,  drop_out_fc:.5, drop_out_conv:.7})\n",
    "        \n",
    "        #evaluate training data   \n",
    "        train_loss, train_accuracy = evaluate(X_train, y_train)\n",
    "        train_loss_all.append(train_loss)\n",
    "        train_acc_all.append(train_accuracy)\n",
    "\n",
    "        #evaluate Validation Data\n",
    "        valid_loss, valid_accuracy = evaluate(X_valid, y_valid)\n",
    "        valid_loss_all.append(valid_loss)\n",
    "        valid_acc_all.append(valid_accuracy)\n",
    "        print(\"EPOCH {}, T Loss: {:.3f}, V Loss: {:.3f}, T Acc: {:.3f}, V Acc: {:.3f}\"\\\n",
    "              .format(i+1, train_loss, valid_loss, train_accuracy, valid_accuracy))\n",
    "        \n",
    "        if valid_accuracy > best_valid_acc:\n",
    "            best_valid_acc = valid_accuracy\n",
    "            saver.save(sess, model_name)\n",
    "    print(\"Model saved with best validation accuracy of: {:.3f}\".format(best_valid_acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3I0xJgBZuBtjTAD6/wHPncXCX+JLCwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2855f5e908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUsoLabgrpZQX+v8BBzILc7al5mcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f28555c1e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_loss_all, 'b', label='Training Loss')\n",
    "plt.plot(valid_loss_all, 'g', label='Validation Loss')\n",
    "plt.legend(loc=1)\n",
    "plt.title('Loss Graph')\n",
    "plt.show()\n",
    "\n",
    "plt.plot(train_acc_all, 'b', label='Training Accuracy')\n",
    "plt.plot(valid_acc_all, 'g', label='Validation Accuracy')\n",
    "plt.legend(loc=4)\n",
    "plt.title('Accuracy Graph')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.166, Test Accuracy = 0.969\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the Test Data\n",
    "saver = tf.train.Saver()\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "\n",
    "    test_loss, test_accuracy = evaluate(X_test, y_test)\n",
    "    print(\"Test loss: {:.3F}, Test Accuracy = {:.3f}\".format(test_loss, test_accuracy))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Step 3: Test a Model on New Images\n",
    "\n",
    "To give yourself more insight into how your model is working, download at least five pictures of German traffic signs from the web and use your model to predict the traffic sign type.\n",
    "\n",
    "You may find `signnames.csv` useful as it contains mappings from the class id (integer) to the actual sign name."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and Output the Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#load images\n",
    "\n",
    "import os\n",
    "import cv2\n",
    "\n",
    "def load_images(path, new_size=32):\n",
    "    images = []\n",
    "    for file in os.listdir(path):\n",
    "        if not file.startswith('.'):\n",
    "            #print(file)\n",
    "            image = cv2.imread(os.path.join(path, file))\n",
    "            image = cv2.resize(image, (new_size, new_size)) \n",
    "            image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "            images.append(image)\n",
    "    return images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done\n"
     ]
    }
   ],
   "source": [
    "path = r'./images/'\n",
    "xx_test = np.asarray(load_images(path, new_size=32))\n",
    "yy_test = [1, 24, 38, 12, 31, 23, 5, 22, 15, 14, 40, 14, 33, 13, 13, 12, 33, 35, 18, 18, 34]\n",
    "\n",
    "actual_test_images = xx_test\n",
    "# normalize and grayscale\n",
    "xx_test = convert_to_grayscale(xx_test)\n",
    "xx_test = normalize(xx_test)\n",
    "print('done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "actual images in 1st row and grayscaled and normalized images in 2nd row\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABLkAAABiCAYAAAC8jQEyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQRAEoewRI5cgCIIgCIIgCIIgCIJQ9vw/IZacrjE8PCsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f285c028828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABLkAAABiCAYAAAC8jQEyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nW3gb8jA5AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f285c0289e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Load  images\n",
    "print('actual images in 1st row and grayscaled and normalized images in 2nd row')\n",
    "print_images(actual_test_images, yy_test, row=1, col=len(yy_test), cmap=None)\n",
    "print_images(xx_test, yy_test, row=1, col=len(yy_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predict the Sign Type for Each Image and find the Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy = 0.952\n"
     ]
    }
   ],
   "source": [
    "### Run the predictions here and use the model to output the prediction for each image.\n",
    "### Make sure to pre-process the images with the same pre-processing pipeline used earlier.\n",
    "### Feel free to use as many code cells as needed.\n",
    "#Evaluate\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "\n",
    "    _, test_accuracy = evaluate(xx_test, yy_test)\n",
    "    print(\"Test Accuracy = {:.3f}\".format(test_accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "### Calculate the accuracy for these 5 new images. \n",
    "### For example, if the model predicted 1 out of 5 signs correctly, it's 20% accurate on these new images.\n",
    "softmax_logits = tf.nn.softmax(logits)\n",
    "top_k = tf.nn.top_k(softmax_logits, k=3)\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "\n",
    "    tp_k = sess.run(top_k, feed_dict={x: xx_test, drop_out_fc:1., drop_out_conv:1.})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Output Top Softmax Probabilities For Each Image Found on the Web"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "float_formatter = lambda x: \"%03.0f\" % (x * 100)\n",
    "np.set_printoptions(formatter={'float_kind':float_formatter})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#  Prediction (%) ->   1, 2, 3 predictions\n",
      "--------------------------------------------------\n",
      "01 [100 000 000]  ->  [1 2 5]\n",
      "02 [100 000 000]  ->  [24 27 28]\n",
      "03 [100 000 000]  ->  [38 17 13]\n",
      "04 [100 000 000]  ->  [12 40  2]\n",
      "05 [100 000 000]  ->  [31 21 25]\n",
      "06 [100 000 000]  ->  [23 19 20]\n",
      "07 [097 002 000]  ->  [5 2 3]\n",
      "08 [100 000 000]  ->  [22 25 29]\n",
      "09 [100 000 000]  ->  [15 12 35]\n",
      "10 [100 000 000]  ->  [14  1  2]\n",
      "11 [100 000 000]  ->  [40 12  7]\n",
      "12 [100 000 000]  ->  [14  1  2]\n",
      "13 [100 000 000]  ->  [33 35 37]\n",
      "14 [100 000 000]  ->  [13  3  2]\n",
      "15 [100 000 000]  ->  [13  0  1]\n",
      "16 [100 000 000]  ->  [12 40  0]\n",
      "17 [061 027 011]  ->  [20 26 18]\n",
      "18 [100 000 000]  ->  [35 33  3]\n",
      "19 [100 000 000]  ->  [18 26 27]\n",
      "20 [100 000 000]  ->  [18 26 27]\n",
      "21 [100 000 000]  ->  [34 38 35]\n",
      "\n",
      "wrongly identifid images:\n",
      "Image wrongly identified as:  20 while actual image is:  33\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAFwAAABZCAYAAACzIkPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxr9x4NiG7ToAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2855616240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Print out the top softmax probabilities for the predictions on the German traffic sign images found on the web. \n",
    "### Feel free to use as many code cells as needed.\n",
    "print('#  Prediction (%) ->   1, 2, 3 predictions')\n",
    "print('-'*50)\n",
    "for x in range(len(tp_k[0])):\n",
    "    print('{:02d} {}  ->  {}'.format(x+1, tp_k[0][x], tp_k[1][x]))\n",
    "\n",
    "print('\\nwrongly identifid images:')\n",
    "wrong_images = []\n",
    "for i in range(len(yy_test)):\n",
    "    if tp_k[1][i][0] != yy_test[i]:\n",
    "        print('Image wrongly identified as: ', tp_k[1][i][0], 'while actual image is: ', yy_test[i])\n",
    "        wrong_images.append(actual_test_images[i])\n",
    "print_images(wrong_images, None, row=1, col=len(wrong_images))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Take Away\n",
    "As can be seen from the predictions, the model got confused with one images, out of 21 total images.   \n",
    "This can further be reduced by increasing the training data using Augmentation. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Project Writeup\n",
    "\n",
    "Once you have completed the code implementation, document your results in a project writeup using this [template](https://github.com/udacity/CarND-Traffic-Sign-Classifier-Project/blob/master/writeup_template.md) as a guide. The writeup can be in a markdown or pdf file. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **Note**: Once you have completed all of the code implementations and successfully answered each question above, you may finalize your work by exporting the iPython Notebook as an HTML document. You can do this by using the menu above and navigating to  \\n\",\n",
    "    \"**File -> Download as -> HTML (.html)**. Include the finished document along with this notebook as your submission."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bibliography\n",
    "Preprocessing and Data Augmentiation: https://benanne.github.io/2015/03/17/plankton.html  \n",
    "https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html  "
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  },
  "widgets": {
   "state": {},
   "version": "1.1.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}