Traffic_Sign_Classifier.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": {},
   "outputs": [],
   "source": [
    "# Load pickled data\n",
    "import pickle\n",
    "\n",
    "# TODO: Fill this in based on where you saved the training and testing data\n",
    "\n",
    "training_file = '../data/train.p'\n",
    "validation_file= '../data/valid.p'\n",
    "testing_file = '../data/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": "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": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training examples = 34799\n",
      "Number of validating examples = 4410\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(set(y_valid))\n",
    "\n",
    "print(\"Number of training examples =\", n_train)\n",
    "print(\"Number of validating examples =\", n_validation)\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": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f161971cba8>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Data exploration visualization code goes here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "import numpy as np \n",
    "import cv2\n",
    "\n",
    "# Visualizations will be shown in the notebook.\n",
    "%matplotlib inline\n",
    "\n",
    "fig=plt.figure(figsize=(16, 8))\n",
    "\n",
    "columns = 8\n",
    "rows = 5\n",
    "\n",
    "for i in range(1, columns*rows +1):\n",
    "    index = random.randint(0, len(X_train))\n",
    "    image = X_train[index]\n",
    "    fig.add_subplot(rows, columns, i)\n",
    "    plt.imshow(image)\n",
    "    \n",
    "plt.show()   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1628b8ab00>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Distribution of classes in the training\n",
    "from collections import Counter \n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "def classes_dist(y):\n",
    "    d = Counter(y)\n",
    "    df = pd.DataFrame.from_dict(d, orient='index').reset_index()\n",
    "    df = df.rename(columns={'index':'class', 0:'number_of_instances'})\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=(16, 5))\n",
    "    sns.barplot(x=\"class\", y=\"number_of_instances\", data=df, palette=\"Blues_d\", ax=ax)\n",
    "    plt.show()\n",
    "    \n",
    "classes_dist(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Augmenting the training data by rotating, shifting and changing colors\n",
    "### for classes, whose number of training instances is less than mean\n",
    "def random_rotation(img):\n",
    "    rot = [15, -15]\n",
    "    i = random.randint(0, 1)\n",
    "    M = cv2.getRotationMatrix2D((img.shape[1]/2, img.shape[0]/2), rot[i], 1.0)\n",
    "    return cv2.warpAffine(img, M , (img.shape[1], img.shape[0]))\n",
    "\n",
    "def random_noise(img):\n",
    "    gaussian_noise = img.copy()\n",
    "    cv2.randn(gaussian_noise, 0, 50)\n",
    "    return img + gaussian_noise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.utils import shuffle\n",
    "\n",
    "\n",
    "d = Counter(y_train)\n",
    "mean = int(np.mean(list(d.values())))\n",
    "\n",
    "transformations = {\n",
    "    'rotate': random_rotation,\n",
    "    'noise': random_noise\n",
    "}\n",
    "\n",
    "for k, v in d.items():\n",
    "    # if the total number of instances for this particular class less than mean \n",
    "    # we expand the training data set via augmenting the existing images\n",
    "    \n",
    "    if v < mean: \n",
    "        samples = [X_train[i] for i,y in enumerate(y_train) if y == k]\n",
    "        \n",
    "        ## apply augmentation\n",
    "        x_list= [random_rotation(x) for x in samples]\n",
    "        #x_list = np.append(x_list, [flip(x) for x in samples], axis=0)\n",
    "        x_list = np.append(x_list, [random_noise(x) for x in samples], axis=0)\n",
    "        \n",
    "        x_list = shuffle(x_list)\n",
    "        \n",
    "        amount = mean-v if mean-v < len(x_list) else len(x_list)\n",
    "        X_train = np.append(X_train, x_list[: amount], axis=0)\n",
    "        y_train = np.append(y_train, amount*[k], axis=0)                 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f15dfedba90>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# check the distribution after augmenting\n",
    "\n",
    "classes_dist(y_train)"
   ]
  },
  {
   "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": 8,
   "metadata": {},
   "outputs": [],
   "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."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f15ba1000f0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f15ba188b00>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.utils import shuffle\n",
    "X_train, y_train = shuffle(X_train, y_train)\n",
    "plt.imshow(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array([cv2.cvtColor(x, cv2.COLOR_RGB2GRAY) for x in X_train])\n",
    "X_valid = np.array([cv2.cvtColor(x, cv2.COLOR_RGB2GRAY) for x in X_valid])\n",
    "X_test = np.array([cv2.cvtColor(x, cv2.COLOR_RGB2GRAY) for x in X_test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image data shape = (32, 32)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f15c22b5fd0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f15dff22438>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_shape = X_train[0].shape\n",
    "print(\"Image data shape =\", image_shape)\n",
    "plt.imshow(X_train[0], cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array([np.expand_dims(cv2.normalize(x,  np.zeros((32, 32, 1)), 0, 255, cv2.NORM_MINMAX), -1) for x in X_train])\n",
    "X_valid = np.array([np.expand_dims(cv2.normalize(x,  np.zeros((32, 32, 1)), 0, 255, cv2.NORM_MINMAX), -1) for x in X_valid])\n",
    "X_test = np.array([np.expand_dims(cv2.normalize(x,  np.zeros((32, 32, 1)), 0, 255, cv2.NORM_MINMAX), -1) for x in X_test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image data shape = (32, 32, 1)\n"
     ]
    }
   ],
   "source": [
    "image_shape = X_train[0].shape\n",
    "print(\"Image data shape =\", image_shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "x = tf.placeholder(tf.float32, (None, 32, 32, 1))\n",
    "y = tf.placeholder(tf.int32, (None))\n",
    "keep_prob = tf.placeholder(tf.float32) # probability to keep units\n",
    "one_hot_y = tf.one_hot(y, 43)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Define your architecture here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "from tensorflow.contrib.layers import flatten\n",
    "\n",
    "def LeNet(x):    \n",
    "    # Arguments used for tf.truncated_normal, randomly defines variables for the weights and biases for each layer\n",
    "    mu = 0\n",
    "    sigma = 0.1\n",
    "    \n",
    "    weights = {\n",
    "        'wc1': tf.Variable(tf.truncated_normal([5, 5, 1, 6], mean=mu, stddev=sigma)),\n",
    "        'wc2': tf.Variable(tf.truncated_normal([5, 5, 6, 16], mean=mu, stddev=sigma)),\n",
    "        'wd2': tf.Variable(tf.truncated_normal([400, 120],  mean=mu, stddev=sigma)),\n",
    "        'wd3': tf.Variable(tf.truncated_normal([120, 84],  mean=mu, stddev=sigma)),\n",
    "        'out': tf.Variable(tf.truncated_normal([84, 43], mean=mu, stddev=sigma))}\n",
    "\n",
    "    biases = {\n",
    "        'bc1': tf.Variable(tf.truncated_normal([6], mean=mu, stddev=sigma)),\n",
    "        'bc2': tf.Variable(tf.random_normal([16],  mean=mu, stddev=sigma)),\n",
    "        'bd2': tf.Variable(tf.random_normal([120],  mean=mu, stddev=sigma)),\n",
    "        'bd3': tf.Variable(tf.random_normal([84],  mean=mu, stddev=sigma)),\n",
    "        'out': tf.Variable(tf.random_normal([43], mean=mu, stddev=sigma))}\n",
    "    \n",
    "    # TODO: Layer 1: Convolutional. Input = 32x32x1. Output = 28x28x6.\n",
    "    conv1 = tf.nn.conv2d(x, weights['wc1'], strides=[1, 1, 1, 1], padding='VALID')\n",
    "    conv1 = tf.nn.bias_add(conv1, biases['bc1'])\n",
    "    \n",
    "    # TODO: Activation.\n",
    "    conv1 = tf.nn.relu(conv1)\n",
    "    \n",
    "    # TODO: Pooling. Input = 28x28x6. Output = 14x14x6.\n",
    "    conv1 =  tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "    fc0 = flatten(conv1)\n",
    "\n",
    "    # TODO: Layer 2: Convolutional. Output = 10x10x16.\n",
    "    conv2 = tf.nn.conv2d(conv1, weights['wc2'], strides=[1, 1, 1, 1], padding='VALID')\n",
    "    conv2 = tf.nn.bias_add(conv2, biases['bc2'])\n",
    "    \n",
    "    # TODO: Activation.\n",
    "    conv2 = tf.nn.relu(conv2)\n",
    "\n",
    "    # TODO: Pooling. Input = 10x10x16. Output = 5x5x16.\n",
    "    conv2 =  tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "    conv2 = tf.nn.dropout(conv2, keep_prob)\n",
    "\n",
    "    # TODO: Flatten. Input = 5x5x16. Output = 400.\n",
    "    fc1   = flatten(conv2)\n",
    "    \n",
    "    # TODO: Layer 3: Fully Connected. Input = 400. Output = 120.\n",
    "    fc2 = tf.add(tf.matmul(fc1, weights['wd2']), biases['bd2'])  \n",
    "    \n",
    "    # TODO: Activation.\n",
    "    fc2 = tf.nn.relu(fc2)\n",
    "    fc2 = tf.nn.dropout(fc2, keep_prob)\n",
    "\n",
    "    # TODO: Layer 4: Fully Connected. Input = 120. Output = 84.\n",
    "    fc3 = tf.add(tf.matmul(fc2, weights['wd3']), biases['bd3'])  \n",
    "    \n",
    "    # TODO: Activation.\n",
    "    fc3 = tf.nn.relu(fc3)\n",
    "    \n",
    "    # TODO: Layer 5: Fully Connected. Input = 84. Output = 10.\n",
    "    #logits = tf.add(tf.matmul(tf.concat([fc0, fc3], 1), weights['out']), biases['out'])\n",
    "    logits = tf.add(tf.matmul(fc3, weights['out']), biases['out'])\n",
    "    \n",
    "    \n",
    "    return logits\n"
   ]
  },
  {
   "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": {},
   "outputs": [],
   "source": [
    "### Train your model here.\n",
    "### Calculate and report the accuracy on the training and validation set.\n",
    "### Once a final model architecture is selected, \n",
    "### the accuracy on the test set should be calculated and reported as well.\n",
    "### Feel free to use as many code cells as needed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "rate = 0.001\n",
    "\n",
    "EPOCHS = 100\n",
    "BATCH_SIZE = 128\n",
    "\n",
    "logits = LeNet(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": "code",
   "execution_count": 18,
   "metadata": {},
   "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",
    "saver = tf.train.Saver()\n",
    "\n",
    "def evaluate(X_data, y_data):\n",
    "    num_examples = len(X_data)\n",
    "    total_accuracy = 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",
    "        accuracy = sess.run(accuracy_operation, feed_dict={x: batch_x, y: batch_y, keep_prob: 1})\n",
    "        total_accuracy += (accuracy * len(batch_x))\n",
    "    return total_accuracy / num_examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "\n",
      "EPOCH 1 ...\n",
      "Training Accuracy = 0.107\n",
      "Validation Accuracy = 0.128\n",
      "\n",
      "EPOCH 2 ...\n",
      "Training Accuracy = 0.195\n",
      "Validation Accuracy = 0.237\n",
      "\n",
      "EPOCH 3 ...\n",
      "Training Accuracy = 0.276\n",
      "Validation Accuracy = 0.319\n",
      "\n",
      "EPOCH 4 ...\n",
      "Training Accuracy = 0.381\n",
      "Validation Accuracy = 0.441\n",
      "\n",
      "EPOCH 5 ...\n",
      "Training Accuracy = 0.483\n",
      "Validation Accuracy = 0.503\n",
      "\n",
      "EPOCH 6 ...\n",
      "Training Accuracy = 0.551\n",
      "Validation Accuracy = 0.566\n",
      "\n",
      "EPOCH 7 ...\n",
      "Training Accuracy = 0.625\n",
      "Validation Accuracy = 0.624\n",
      "\n",
      "EPOCH 8 ...\n",
      "Training Accuracy = 0.707\n",
      "Validation Accuracy = 0.701\n",
      "\n",
      "EPOCH 9 ...\n",
      "Training Accuracy = 0.754\n",
      "Validation Accuracy = 0.741\n",
      "\n",
      "EPOCH 10 ...\n",
      "Training Accuracy = 0.813\n",
      "Validation Accuracy = 0.803\n",
      "\n",
      "EPOCH 11 ...\n",
      "Training Accuracy = 0.839\n",
      "Validation Accuracy = 0.830\n",
      "\n",
      "EPOCH 12 ...\n",
      "Training Accuracy = 0.869\n",
      "Validation Accuracy = 0.869\n",
      "\n",
      "EPOCH 13 ...\n",
      "Training Accuracy = 0.883\n",
      "Validation Accuracy = 0.872\n",
      "\n",
      "EPOCH 14 ...\n",
      "Training Accuracy = 0.894\n",
      "Validation Accuracy = 0.889\n",
      "\n",
      "EPOCH 15 ...\n",
      "Training Accuracy = 0.908\n",
      "Validation Accuracy = 0.895\n",
      "\n",
      "EPOCH 16 ...\n",
      "Training Accuracy = 0.920\n",
      "Validation Accuracy = 0.904\n",
      "\n",
      "EPOCH 17 ...\n",
      "Training Accuracy = 0.925\n",
      "Validation Accuracy = 0.919\n",
      "\n",
      "EPOCH 18 ...\n",
      "Training Accuracy = 0.930\n",
      "Validation Accuracy = 0.909\n",
      "\n",
      "EPOCH 19 ...\n",
      "Training Accuracy = 0.936\n",
      "Validation Accuracy = 0.915\n",
      "\n",
      "EPOCH 20 ...\n",
      "Training Accuracy = 0.942\n",
      "Validation Accuracy = 0.920\n",
      "\n",
      "EPOCH 21 ...\n",
      "Training Accuracy = 0.940\n",
      "Validation Accuracy = 0.915\n",
      "\n",
      "EPOCH 22 ...\n",
      "Training Accuracy = 0.946\n",
      "Validation Accuracy = 0.927\n",
      "\n",
      "EPOCH 23 ...\n",
      "Training Accuracy = 0.948\n",
      "Validation Accuracy = 0.923\n",
      "\n",
      "EPOCH 24 ...\n",
      "Training Accuracy = 0.957\n",
      "Validation Accuracy = 0.931\n",
      "\n",
      "EPOCH 25 ...\n",
      "Training Accuracy = 0.960\n",
      "Validation Accuracy = 0.931\n",
      "\n",
      "EPOCH 26 ...\n",
      "Training Accuracy = 0.963\n",
      "Validation Accuracy = 0.937\n",
      "\n",
      "EPOCH 27 ...\n",
      "Training Accuracy = 0.958\n",
      "Validation Accuracy = 0.931\n",
      "\n",
      "EPOCH 28 ...\n",
      "Training Accuracy = 0.960\n",
      "Validation Accuracy = 0.929\n",
      "\n",
      "EPOCH 29 ...\n",
      "Training Accuracy = 0.967\n",
      "Validation Accuracy = 0.939\n",
      "\n",
      "EPOCH 30 ...\n",
      "Training Accuracy = 0.972\n",
      "Validation Accuracy = 0.942\n",
      "\n",
      "EPOCH 31 ...\n",
      "Training Accuracy = 0.964\n",
      "Validation Accuracy = 0.927\n",
      "\n",
      "EPOCH 32 ...\n",
      "Training Accuracy = 0.972\n",
      "Validation Accuracy = 0.945\n",
      "\n",
      "EPOCH 33 ...\n",
      "Training Accuracy = 0.973\n",
      "Validation Accuracy = 0.940\n",
      "\n",
      "EPOCH 34 ...\n",
      "Training Accuracy = 0.975\n",
      "Validation Accuracy = 0.947\n",
      "\n",
      "EPOCH 35 ...\n",
      "Training Accuracy = 0.971\n",
      "Validation Accuracy = 0.944\n",
      "\n",
      "EPOCH 36 ...\n",
      "Training Accuracy = 0.973\n",
      "Validation Accuracy = 0.937\n",
      "\n",
      "EPOCH 37 ...\n",
      "Training Accuracy = 0.977\n",
      "Validation Accuracy = 0.940\n",
      "\n",
      "EPOCH 38 ...\n",
      "Training Accuracy = 0.972\n",
      "Validation Accuracy = 0.941\n",
      "\n",
      "EPOCH 39 ...\n",
      "Training Accuracy = 0.973\n",
      "Validation Accuracy = 0.943\n",
      "\n",
      "EPOCH 40 ...\n",
      "Training Accuracy = 0.974\n",
      "Validation Accuracy = 0.934\n",
      "\n",
      "EPOCH 41 ...\n",
      "Training Accuracy = 0.978\n",
      "Validation Accuracy = 0.946\n",
      "\n",
      "EPOCH 42 ...\n",
      "Training Accuracy = 0.978\n",
      "Validation Accuracy = 0.944\n",
      "\n",
      "EPOCH 43 ...\n",
      "Training Accuracy = 0.978\n",
      "Validation Accuracy = 0.951\n",
      "\n",
      "EPOCH 44 ...\n",
      "Training Accuracy = 0.977\n",
      "Validation Accuracy = 0.956\n",
      "\n",
      "EPOCH 45 ...\n",
      "Training Accuracy = 0.981\n",
      "Validation Accuracy = 0.952\n",
      "\n",
      "EPOCH 46 ...\n",
      "Training Accuracy = 0.980\n",
      "Validation Accuracy = 0.954\n",
      "\n",
      "EPOCH 47 ...\n",
      "Training Accuracy = 0.980\n",
      "Validation Accuracy = 0.952\n",
      "\n",
      "EPOCH 48 ...\n",
      "Training Accuracy = 0.982\n",
      "Validation Accuracy = 0.952\n",
      "\n",
      "EPOCH 49 ...\n",
      "Training Accuracy = 0.979\n",
      "Validation Accuracy = 0.951\n",
      "\n",
      "EPOCH 50 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.960\n",
      "\n",
      "EPOCH 51 ...\n",
      "Training Accuracy = 0.982\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 52 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.965\n",
      "\n",
      "EPOCH 53 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 54 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 55 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.952\n",
      "\n",
      "EPOCH 56 ...\n",
      "Training Accuracy = 0.982\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 57 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 58 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.960\n",
      "\n",
      "EPOCH 59 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.952\n",
      "\n",
      "EPOCH 60 ...\n",
      "Training Accuracy = 0.982\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 61 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.958\n",
      "\n",
      "EPOCH 62 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.955\n",
      "\n",
      "EPOCH 63 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 64 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.965\n",
      "\n",
      "EPOCH 65 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.958\n",
      "\n",
      "EPOCH 66 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.958\n",
      "\n",
      "EPOCH 67 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.962\n",
      "\n",
      "EPOCH 68 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.966\n",
      "\n",
      "EPOCH 69 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 70 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 71 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 72 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.963\n",
      "\n",
      "EPOCH 73 ...\n",
      "Training Accuracy = 0.983\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 74 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.965\n",
      "\n",
      "EPOCH 75 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 76 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.966\n",
      "\n",
      "EPOCH 77 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.969\n",
      "\n",
      "EPOCH 78 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.955\n",
      "\n",
      "EPOCH 79 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.962\n",
      "\n",
      "EPOCH 80 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 81 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 82 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.963\n",
      "\n",
      "EPOCH 83 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.955\n",
      "\n",
      "EPOCH 84 ...\n",
      "Training Accuracy = 0.984\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 85 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 86 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.963\n",
      "\n",
      "EPOCH 87 ...\n",
      "Training Accuracy = 0.985\n",
      "Validation Accuracy = 0.958\n",
      "\n",
      "EPOCH 88 ...\n",
      "Training Accuracy = 0.988\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 89 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.967\n",
      "\n",
      "EPOCH 90 ...\n",
      "Training Accuracy = 0.986\n",
      "Validation Accuracy = 0.960\n",
      "\n",
      "EPOCH 91 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 92 ...\n",
      "Training Accuracy = 0.988\n",
      "Validation Accuracy = 0.964\n",
      "\n",
      "EPOCH 93 ...\n",
      "Training Accuracy = 0.989\n",
      "Validation Accuracy = 0.971\n",
      "\n",
      "EPOCH 94 ...\n",
      "Training Accuracy = 0.989\n",
      "Validation Accuracy = 0.966\n",
      "\n",
      "EPOCH 95 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.966\n",
      "\n",
      "EPOCH 96 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.964\n",
      "\n",
      "EPOCH 97 ...\n",
      "Training Accuracy = 0.988\n",
      "Validation Accuracy = 0.965\n",
      "\n",
      "EPOCH 98 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.961\n",
      "\n",
      "EPOCH 99 ...\n",
      "Training Accuracy = 0.987\n",
      "Validation Accuracy = 0.963\n",
      "\n",
      "EPOCH 100 ...\n",
      "Training Accuracy = 0.989\n",
      "Validation Accuracy = 0.968\n",
      "\n",
      "Model saved\n"
     ]
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    num_examples = len(X_train)\n",
    "    \n",
    "    print(\"Training...\")\n",
    "    print()\n",
    "    for i in range(EPOCHS):\n",
    "        X_train, y_train = shuffle(X_train, y_train)\n",
    "        for offset in range(0, num_examples, 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, keep_prob: 0.5})\n",
    "            \n",
    "        training_accuracy = evaluate(X_train, y_train)\n",
    "        validation_accuracy = evaluate(X_valid, y_valid)\n",
    "        print(\"EPOCH {} ...\".format(i+1))\n",
    "        print(\"Training Accuracy = {:.3f}\".format(training_accuracy))\n",
    "        print(\"Validation Accuracy = {:.3f}\".format(validation_accuracy))\n",
    "        print()\n",
    "        \n",
    "    saver.save(sess, './lenet')\n",
    "    print(\"Model saved\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./lenet\n",
      "Test Accuracy = 0.952\n"
     ]
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "\n",
    "    test_accuracy = evaluate(X_test, y_test)\n",
    "    print(\"Test Accuracy = {:.3f}\".format(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": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 32, 3)\n",
      "(32, 32, 3)\n",
      "(32, 32, 3)\n",
      "(32, 32, 3)\n",
      "(32, 32, 3)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f15bf803c50>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Load the images and plot them here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import glob\n",
    "\n",
    "test_images = glob.glob('examples/sample*.png')\n",
    "fig=plt.figure(figsize=(16, 2))\n",
    "x_test = []\n",
    "\n",
    "for i, img_path in enumerate(test_images):\n",
    "    img = cv2.imread(img_path) \n",
    "    x_test.append(img)\n",
    "    fig.add_subplot(1, 5, i+1)\n",
    "    plt.imshow(img)\n",
    "    print(img.shape)\n",
    "    \n",
    "plt.show() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predict the Sign Type for Each Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image data shape = (32, 32, 1)\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",
    "x_test = np.array([cv2.cvtColor(x, cv2.COLOR_RGB2GRAY) for x in x_test])\n",
    "x_test = np.array([np.expand_dims(cv2.normalize(x,  np.zeros((32, 32)), 0, 255, cv2.NORM_MINMAX), -1) for x in x_test])\n",
    "\n",
    "image_shape = x_test[0].shape\n",
    "print(\"Image data shape =\", image_shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "yy_test = [1, 20, 36, 28, 14]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./lenet\n",
      "Test Accuracy = 1.000\n"
     ]
    }
   ],
   "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",
    "\n",
    "### Print out the top five 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",
    "\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "    \n",
    "    test_accuracy = evaluate(x_test, yy_test)\n",
    "    print(\"Test Accuracy = {:.3f}\".format(test_accuracy))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Output Top 5 Softmax Probabilities For Each Image Found on the Web"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each of the new images, print out the model's softmax probabilities to show the **certainty** of the model's predictions (limit the output to the top 5 probabilities for each image). [`tf.nn.top_k`](https://www.tensorflow.org/versions/r0.12/api_docs/python/nn.html#top_k) could prove helpful here. \n",
    "\n",
    "The example below demonstrates how tf.nn.top_k can be used to find the top k predictions for each image.\n",
    "\n",
    "`tf.nn.top_k` will return the values and indices (class ids) of the top k predictions. So if k=3, for each sign, it'll return the 3 largest probabilities (out of a possible 43) and the correspoding class ids.\n",
    "\n",
    "Take this numpy array as an example. The values in the array represent predictions. The array contains softmax probabilities for five candidate images with six possible classes. `tf.nn.top_k` is used to choose the three classes with the highest probability:\n",
    "\n",
    "```\n",
    "# (5, 6) array\n",
    "a = np.array([[ 0.24879643,  0.07032244,  0.12641572,  0.34763842,  0.07893497,\n",
    "         0.12789202],\n",
    "       [ 0.28086119,  0.27569815,  0.08594638,  0.0178669 ,  0.18063401,\n",
    "         0.15899337],\n",
    "       [ 0.26076848,  0.23664738,  0.08020603,  0.07001922,  0.1134371 ,\n",
    "         0.23892179],\n",
    "       [ 0.11943333,  0.29198961,  0.02605103,  0.26234032,  0.1351348 ,\n",
    "         0.16505091],\n",
    "       [ 0.09561176,  0.34396535,  0.0643941 ,  0.16240774,  0.24206137,\n",
    "         0.09155967]])\n",
    "```\n",
    "\n",
    "Running it through `sess.run(tf.nn.top_k(tf.constant(a), k=3))` produces:\n",
    "\n",
    "```\n",
    "TopKV2(values=array([[ 0.34763842,  0.24879643,  0.12789202],\n",
    "       [ 0.28086119,  0.27569815,  0.18063401],\n",
    "       [ 0.26076848,  0.23892179,  0.23664738],\n",
    "       [ 0.29198961,  0.26234032,  0.16505091],\n",
    "       [ 0.34396535,  0.24206137,  0.16240774]]), indices=array([[3, 0, 5],\n",
    "       [0, 1, 4],\n",
    "       [0, 5, 1],\n",
    "       [1, 3, 5],\n",
    "       [1, 4, 3]], dtype=int32))\n",
    "```\n",
    "\n",
    "Looking just at the first row we get `[ 0.34763842,  0.24879643,  0.12789202]`, you can confirm these are the 3 largest probabilities in `a`. You'll also notice `[3, 0, 5]` are the corresponding indices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./lenet\n",
      "TopKV2(values=array([[  9.99997139e-01,   2.51626830e-06,   2.70898397e-07,\n",
      "          1.30178819e-07,   1.32103608e-08],\n",
      "       [  6.93328023e-01,   1.57814413e-01,   6.91553578e-02,\n",
      "          2.75372881e-02,   2.00350378e-02],\n",
      "       [  9.99946594e-01,   3.68386682e-05,   1.36604613e-05,\n",
      "          1.19015715e-06,   9.58731675e-07],\n",
      "       [  7.11617291e-01,   2.34091803e-01,   1.87694058e-02,\n",
      "          1.23576773e-02,   8.07854719e-03],\n",
      "       [  9.98848081e-01,   3.42276267e-04,   3.07735492e-04,\n",
      "          1.83745506e-04,   8.74467078e-05]], dtype=float32), indices=array([[ 1,  2,  0,  6,  7],\n",
      "       [20, 28, 18, 29, 30],\n",
      "       [36, 35, 33, 14, 34],\n",
      "       [28, 20, 23, 18, 30],\n",
      "       [14, 38, 33, 17, 10]], dtype=int32))\n"
     ]
    }
   ],
   "source": [
    "### Print out the top five 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",
    "top_predications =  tf.nn.top_k(tf.nn.softmax(logits), k = 5)\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, tf.train.latest_checkpoint('.'))\n",
    "    \n",
    "    res = sess.run(top_predications, feed_dict={x: x_test, keep_prob:1.0})\n",
    "    print(res)"
   ]
  },
  {
   "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": [
    "---\n",
    "\n",
    "## Step 4 (Optional): Visualize the Neural Network's State with Test Images\n",
    "\n",
    " This Section is not required to complete but acts as an additional excersise for understaning the output of a neural network's weights. While neural networks can be a great learning device they are often referred to as a black box. We can understand what the weights of a neural network look like better by plotting their feature maps. After successfully training your neural network you can see what it's feature maps look like by plotting the output of the network's weight layers in response to a test stimuli image. From these plotted feature maps, it's possible to see what characteristics of an image the network finds interesting. For a sign, maybe the inner network feature maps react with high activation to the sign's boundary outline or to the contrast in the sign's painted symbol.\n",
    "\n",
    " Provided for you below is the function code that allows you to get the visualization output of any tensorflow weight layer you want. The inputs to the function should be a stimuli image, one used during training or a new one you provided, and then the tensorflow variable name that represents the layer's state during the training process, for instance if you wanted to see what the [LeNet lab's](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) feature maps looked like for it's second convolutional layer you could enter conv2 as the tf_activation variable.\n",
    "\n",
    "For an example of what feature map outputs look like, check out NVIDIA's results in their paper [End-to-End Deep Learning for Self-Driving Cars](https://devblogs.nvidia.com/parallelforall/deep-learning-self-driving-cars/) in the section Visualization of internal CNN State. NVIDIA was able to show that their network's inner weights had high activations to road boundary lines by comparing feature maps from an image with a clear path to one without. Try experimenting with a similar test to show that your trained network's weights are looking for interesting features, whether it's looking at differences in feature maps from images with or without a sign, or even what feature maps look like in a trained network vs a completely untrained one on the same sign image.\n",
    "\n",
    "<figure>\n",
    " <img src=\"visualize_cnn.png\" width=\"380\" alt=\"Combined Image\" />\n",
    " <figcaption>\n",
    " <p></p> \n",
    " <p style=\"text-align: center;\"> Your output should look something like this (above)</p> \n",
    " </figcaption>\n",
    "</figure>\n",
    " <p></p> \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Visualize your network's feature maps here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "\n",
    "# image_input: the test image being fed into the network to produce the feature maps\n",
    "# tf_activation: should be a tf variable name used during your training procedure that represents the calculated state of a specific weight layer\n",
    "# activation_min/max: can be used to view the activation contrast in more detail, by default matplot sets min and max to the actual min and max values of the output\n",
    "# plt_num: used to plot out multiple different weight feature map sets on the same block, just extend the plt number for each new feature map entry\n",
    "\n",
    "def outputFeatureMap(image_input, tf_activation, activation_min=-1, activation_max=-1 ,plt_num=1):\n",
    "    # Here make sure to preprocess your image_input in a way your network expects\n",
    "    # with size, normalization, ect if needed\n",
    "    # image_input =\n",
    "    # Note: x should be the same name as your network's tensorflow data placeholder variable\n",
    "    # If you get an error tf_activation is not defined it may be having trouble accessing the variable from inside a function\n",
    "    activation = tf_activation.eval(session=sess,feed_dict={x : image_input})\n",
    "    featuremaps = activation.shape[3]\n",
    "    plt.figure(plt_num, figsize=(15,15))\n",
    "    for featuremap in range(featuremaps):\n",
    "        plt.subplot(6,8, featuremap+1) # sets the number of feature maps to show on each row and column\n",
    "        plt.title('FeatureMap ' + str(featuremap)) # displays the feature map number\n",
    "        if activation_min != -1 & activation_max != -1:\n",
    "            plt.imshow(activation[0,:,:, featuremap], interpolation=\"nearest\", vmin =activation_min, vmax=activation_max, cmap=\"gray\")\n",
    "        elif activation_max != -1:\n",
    "            plt.imshow(activation[0,:,:, featuremap], interpolation=\"nearest\", vmax=activation_max, cmap=\"gray\")\n",
    "        elif activation_min !=-1:\n",
    "            plt.imshow(activation[0,:,:, featuremap], interpolation=\"nearest\", vmin=activation_min, cmap=\"gray\")\n",
    "        else:\n",
    "            plt.imshow(activation[0,:,:, featuremap], interpolation=\"nearest\", cmap=\"gray\")"
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}