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": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Import required modules\n",
    "import pickle\n",
    "import numpy as np\n",
    "import cv2\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Step 0: Load The Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data loaded\n",
      "training data size (34799, 32, 32, 3)\n",
      "validation data size (4410, 32, 32, 3)\n",
      "testing data size (12630, 32, 32, 3)\n"
     ]
    }
   ],
   "source": [
    "#Load pickled data\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']\n",
    "\n",
    "assert(len(X_train) == len(y_train))\n",
    "assert(len(X_valid) == len(y_valid))\n",
    "assert(len(X_test) == len(y_test))\n",
    "\n",
    "print(\"data loaded\")\n",
    "print(\"training data size\", X_train.shape)\n",
    "print(\"validation data size\", X_valid.shape)\n",
    "print(\"testing data size\", X_test.shape)"
   ]
  },
  {
   "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": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training examples = 34799\n",
      "Number of validation 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",
    "# Number of training examples\n",
    "n_train = X_train.shape[0]\n",
    "\n",
    "# Number of validation examples\n",
    "n_validation = X_valid.shape[0]\n",
    "\n",
    "# Number of testing examples.\n",
    "n_test = X_test.shape[0]\n",
    "\n",
    "# Shape of an traffic sign image\n",
    "image_shape = X_train.shape[1:]\n",
    "\n",
    "# Unique classes/labels there are in the dataset.\n",
    "n_classes = len(set(y_train))\n",
    "\n",
    "print(\"Number of training examples =\", n_train)\n",
    "print(\"Number of validation 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": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No of classes and labels\n",
      "43 43\n",
      "list containing no of images for each class in training data set\n",
      "[180, 1980, 2010, 1260, 1770, 1650, 360, 1290, 1260, 1320, 1800, 1170, 1890, 1920, 690, 540, 360, 990, 1080, 180, 300, 270, 330, 450, 240, 1350, 540, 210, 480, 240, 390, 690, 210, 599, 360, 1080, 330, 180, 1860, 270, 300, 210, 210]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0bf9d94be0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Data exploration visualization code goes here.\n",
    "\n",
    "\n",
    "#Dictionary data to map label digits to text\n",
    "#First colomn contains digit and second colomn contains the text\n",
    "label_digit_to_text = {}\n",
    "with open('signnames.csv', 'r') as file:\n",
    "    reader = csv.reader(file)\n",
    "    for read in reader:\n",
    "        if read[0].isdigit():\n",
    "            label_digit_to_text[int(read[0])] = read[1]\n",
    "\n",
    "#No of classes\n",
    "class_no = len(label_digit_to_text)\n",
    "    \n",
    "class_data_count = [0 for i in range(class_no)]\n",
    "label_to_image = {}\n",
    "for index, label in enumerate(y_train):\n",
    "    if label not in label_to_image:\n",
    "        label_to_image[label] = X_train[index]\n",
    "    class_data_count[label] = class_data_count[label] + 1\n",
    "    \n",
    "print (\"No of classes and labels\")\n",
    "print (len(class_data_count),len(label_to_image))\n",
    "print (\"list containing no of images for each class in training data set\")\n",
    "print (class_data_count)\n",
    "\n",
    "#lets plot 15 of 43 classes\n",
    "fig,axes = plt.subplots(5, 3, figsize=(15,15))\n",
    "#fig.subplots_adjust(hspace = 0.4,wspace=0.2)\n",
    "for row in range(5):\n",
    "    for col in range(3):\n",
    "        a = axes[row][col]\n",
    "        a.axis('off')\n",
    "        a.imshow(label_to_image[(row*5) + col])\n",
    "        a.set_title(label_digit_to_text[(row*5) + col], fontsize=12)\n",
    "plt.savefig('examples/classes_with_text.jpg')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3 43\n",
      "3 43\n",
      "[[180, 1980, 2010, 1260, 1770, 1650, 360, 1290, 1260, 1320, 1800, 1170, 1890, 1920, 690, 540, 360, 990, 1080, 180, 300, 270, 330, 450, 240, 1350, 540, 210, 480, 240, 390, 690, 210, 599, 360, 1080, 330, 180, 1860, 270, 300, 210, 210], [30, 240, 240, 150, 210, 210, 60, 150, 150, 150, 210, 150, 210, 240, 90, 90, 60, 120, 120, 30, 60, 60, 60, 60, 30, 150, 60, 30, 60, 30, 60, 90, 30, 90, 60, 120, 60, 30, 210, 30, 60, 30, 30], [60, 720, 750, 450, 660, 630, 150, 450, 450, 480, 660, 420, 690, 720, 270, 210, 150, 360, 390, 60, 90, 90, 120, 150, 90, 480, 180, 60, 150, 90, 150, 270, 60, 210, 120, 390, 120, 60, 690, 90, 90, 60, 90]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0bf955bcc0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# lets plot for every class how much data we have in training, validation and testing datasets\n",
    "data_train = []\n",
    "data_valid = []\n",
    "data_test = []\n",
    "for i in range(43):\n",
    "    data_train.append(0)\n",
    "    data_valid.append(0)\n",
    "    data_test.append(0)\n",
    "data = []\n",
    "data.append(data_train)\n",
    "data.append(data_valid)\n",
    "data.append(data_test)\n",
    "print(len(data), len(data[0]));\n",
    "\n",
    "for i in y_train:\n",
    "    data[0][i] = data[0][i] + 1;\n",
    "    \n",
    "for i in y_valid:\n",
    "    data[1][i] = data[1][i] + 1;\n",
    "    \n",
    "for i in y_test:\n",
    "    data[2][i] = data[2][i] + 1;\n",
    "print (len(data), len(data[0]))\n",
    "\n",
    "print(data)\n",
    "# x = np.array(data)\n",
    "# print(x.shape)\n",
    "# print(x)\n",
    "colors = ['red', 'tan', 'lime']\n",
    "#plt.hist(x, 43, density=True, histtype='bar', stacked=True)\n",
    "bars = np.add(data[0], data[1]).tolist()\n",
    "p1 = plt.bar(np.arange(43), data[0], width = 0.8, color='r', edgecolor='white')  # training data set\n",
    "p2 = plt.bar(np.arange(43), data[1], bottom=data[0], width = 0.8, color='k', edgecolor='white') # validation data set\n",
    "p3 = plt.bar(np.arange(43), data[2], bottom=bars, width = 0.8, color='g', edgecolor='white') # testing data set\n",
    "plt.savefig('examples/classes_data_stacked_bar_plot.jpg')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(34799, 32, 32, 1)\n"
     ]
    }
   ],
   "source": [
    "### Preprocess the data here. It is required to normalize the data. Other preprocessing steps could include \n",
    "### converting to grayscale, etc.\n",
    "def process_image(image):\n",
    "    \"\"\"normalize the image by dividing 255.0 and then convert the image to gray scale\"\"\"\n",
    "    norm_image = np.array(image/255.0, dtype=np.float32)\n",
    "    gray_norm_image = cv2.cvtColor(norm_image, cv2.COLOR_RGB2GRAY)\n",
    "    return gray_norm_image.reshape(32,32,1)\n",
    "    \n",
    "features_train = np.array([process_image(image) for image in X_train])\n",
    "labels_train   = y_train\n",
    "\n",
    "features_valid = np.array([process_image(image) for image in X_valid])\n",
    "labels_valid   = y_valid\n",
    "\n",
    "features_test  = np.array([process_image(image) for image in X_test])\n",
    "labels_test    = y_test\n",
    "print(features_train.shape)"
   ]
  },
  {
   "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": "markdown",
   "metadata": {},
   "source": [
    "### Model Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Define your architecture here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import tensorflow as tf\n",
    "from sklearn.utils import shuffle\n",
    "    \n",
    "def traffic_sign_classifier(X):\n",
    "    mu = 0\n",
    "    sigma = 0.1\n",
    "    \n",
    "    ##Layer 1\n",
    "    #convolution layer: 32x32x1 -> 28x28x6\n",
    "    conv1_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 1, 6), mean = mu, stddev = sigma))\n",
    "    conv1_b = tf.Variable(tf.zeros(6))\n",
    "    conv1 = tf.nn.conv2d(X, conv1_W, strides=[1,1,1,1], padding='VALID') + conv1_b\n",
    "    #activation\n",
    "    conv1 = tf.nn.relu(conv1)\n",
    "    #pooling: 28x28x6 -> 14x14x6\n",
    "    conv1 = tf.nn.max_pool(conv1, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')\n",
    "    \n",
    "    ##Layer 2\n",
    "    #convolution layer 14x14x6 -> 10x10x16\n",
    "    conv2_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 6, 16), mean = mu, stddev = sigma))\n",
    "    conv2_b = tf.Variable(tf.zeros(16))\n",
    "    conv2 = tf.nn.conv2d(conv1, conv2_W, strides=[1,1,1,1], padding='VALID') + conv2_b\n",
    "    #activation\n",
    "    conv2 = tf.nn.relu(conv2)\n",
    "    #pooling: 10x10x16 -> 5x5x16\n",
    "    conv2 = tf.nn.max_pool(conv2, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')\n",
    "    \n",
    "    ##Layer 3\n",
    "    #Flatten 5x5x16-> 400\n",
    "    fc0 = tf.contrib.layers.flatten(conv2)\n",
    "    fc0_W = tf.Variable(tf.truncated_normal(shape=(400, 120), mean=mu, stddev = sigma))\n",
    "    fc0_b = tf.Variable(tf.zeros(120))\n",
    "    fc1 = tf.matmul(fc0, fc0_W) + fc0_b\n",
    "    #activation\n",
    "    fc1 = tf.nn.relu(fc1)\n",
    "    \n",
    "    ##Layer 4\n",
    "    fc1_W = tf.Variable(tf.truncated_normal(shape=(120, 84), mean=mu, stddev = sigma))\n",
    "    fc1_b = tf.Variable(tf.zeros(84))\n",
    "    fc2 = tf.matmul(fc1, fc1_W) + fc1_b\n",
    "    #activation\n",
    "    fc2 = tf.nn.relu(fc2)\n",
    "    \n",
    "    ##Layer 5\n",
    "    fc2_W = tf.Variable(tf.truncated_normal(shape=(84, n_classes), mean=mu, stddev = sigma))\n",
    "    fc2_b = tf.Variable(tf.zeros(n_classes))\n",
    "    fc3 = tf.matmul(fc2, fc2_W) + fc2_b\n",
    "    \n",
    "    return fc3\n",
    "\n",
    "\n",
    "# Define training parameters\n",
    "epoch = 500\n",
    "batch_size = 64#128\n",
    "rate = 0.001\n",
    "### Define the model and its input\n",
    "traffic_sign_x = tf.placeholder(tf.float32, (None,32,32,1))\n",
    "traffic_sign_y = tf.placeholder(tf.int32, (None))\n",
    "one_hot_y = tf.one_hot(traffic_sign_y, n_classes)\n",
    "\n",
    "logits = traffic_sign_classifier(traffic_sign_x)\n",
    "\n",
    "# Training pipepline\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",
    "train_operation  = optimizer.minimize(loss_operation)\n",
    "\n",
    "\n",
    "# A function for evaluating the accuracy of a model\n",
    "prediction_operation = tf.equal(tf.argmax(logits, 1), tf.argmax(one_hot_y, 1))\n",
    "accuary_operation = tf.reduce_mean(tf.cast(prediction_operation, tf.float32))\n",
    "saver = tf.train.Saver()\n",
    "model_file = './model'\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(accuary_operation, feed_dict={traffic_sign_x: batch_x, traffic_sign_y: batch_y})\n",
    "        total_accuracy += (accuracy * len(batch_x))\n",
    "    return total_accuracy / num_examples\n",
    "\n",
    "\n",
    "# Starts training\n",
    "def train(X_data, y_data):\n",
    "    with tf.Session() as sess:\n",
    "        sess.run(tf.global_variables_initializer())\n",
    "        print(\"Training ...\")\n",
    "        for i in range(epoch):\n",
    "            #X_train, X_tst, y_train, y_tst = train_test_split(X_data, y_data, test_size=0.1, random_state=int(time.time()))\n",
    "            #begin_time = time.time()\n",
    "            X_data, y_data = shuffle(X_data,y_data)\n",
    "            for offset in range(0, len(X_data), batch_size):\n",
    "                end = offset + batch_size\n",
    "                features, labels = X_data[offset:end], y_data[offset:end]\n",
    "                #print(features.shape,labels.shape)\n",
    "                sess.run(train_operation, feed_dict={traffic_sign_x : features, traffic_sign_y : labels})\n",
    "            validation_accuracy = evaluate(features_valid, labels_valid)\n",
    "            print(\"epoch {} ...\".format(i+1))\n",
    "            print(\"validation accuracy ... = {:.3f}\".format(validation_accuracy))\n",
    "\n",
    "        saver.save(sess, model_file);\n",
    "        print(\"model saved into {}\".format(model_file))"
   ]
  },
  {
   "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": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training ...\n",
      "epoch 1 ...\n",
      "validation accuracy ... = 0.698\n",
      "epoch 2 ...\n",
      "validation accuracy ... = 0.833\n",
      "epoch 3 ...\n",
      "validation accuracy ... = 0.834\n",
      "epoch 4 ...\n",
      "validation accuracy ... = 0.882\n",
      "epoch 5 ...\n",
      "validation accuracy ... = 0.895\n",
      "epoch 6 ...\n",
      "validation accuracy ... = 0.917\n",
      "epoch 7 ...\n",
      "validation accuracy ... = 0.903\n",
      "epoch 8 ...\n",
      "validation accuracy ... = 0.924\n",
      "epoch 9 ...\n",
      "validation accuracy ... = 0.916\n",
      "epoch 10 ...\n",
      "validation accuracy ... = 0.920\n",
      "epoch 11 ...\n",
      "validation accuracy ... = 0.910\n",
      "epoch 12 ...\n",
      "validation accuracy ... = 0.914\n",
      "epoch 13 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 14 ...\n",
      "validation accuracy ... = 0.902\n",
      "epoch 15 ...\n",
      "validation accuracy ... = 0.910\n",
      "epoch 16 ...\n",
      "validation accuracy ... = 0.915\n",
      "epoch 17 ...\n",
      "validation accuracy ... = 0.918\n",
      "epoch 18 ...\n",
      "validation accuracy ... = 0.921\n",
      "epoch 19 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 20 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 21 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 22 ...\n",
      "validation accuracy ... = 0.925\n",
      "epoch 23 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 24 ...\n",
      "validation accuracy ... = 0.930\n",
      "epoch 25 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 26 ...\n",
      "validation accuracy ... = 0.926\n",
      "epoch 27 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 28 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 29 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 30 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 31 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 32 ...\n",
      "validation accuracy ... = 0.917\n",
      "epoch 33 ...\n",
      "validation accuracy ... = 0.917\n",
      "epoch 34 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 35 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 36 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 37 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 38 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 39 ...\n",
      "validation accuracy ... = 0.927\n",
      "epoch 40 ...\n",
      "validation accuracy ... = 0.928\n",
      "epoch 41 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 42 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 43 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 44 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 45 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 46 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 47 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 48 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 49 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 50 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 51 ...\n",
      "validation accuracy ... = 0.916\n",
      "epoch 52 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 53 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 54 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 55 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 56 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 57 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 58 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 59 ...\n",
      "validation accuracy ... = 0.935\n",
      "epoch 60 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 61 ...\n",
      "validation accuracy ... = 0.938\n",
      "epoch 62 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 63 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 64 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 65 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 66 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 67 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 68 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 69 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 70 ...\n",
      "validation accuracy ... = 0.930\n",
      "epoch 71 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 72 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 73 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 74 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 75 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 76 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 77 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 78 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 79 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 80 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 81 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 82 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 83 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 84 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 85 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 86 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 87 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 88 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 89 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 90 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 91 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 92 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 93 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 94 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 95 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 96 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 97 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 98 ...\n",
      "validation accuracy ... = 0.952\n",
      "epoch 99 ...\n",
      "validation accuracy ... = 0.884\n",
      "epoch 100 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 101 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 102 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 103 ...\n",
      "validation accuracy ... = 0.938\n",
      "epoch 104 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 105 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 106 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 107 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 108 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 109 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 110 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 111 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 112 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 113 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 114 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 115 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 116 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 117 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 118 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 119 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 120 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 121 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 122 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 123 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 124 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 125 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 126 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 127 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 128 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 129 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 130 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 131 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 132 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 133 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 134 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 135 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 136 ...\n",
      "validation accuracy ... = 0.952\n",
      "epoch 137 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 138 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 139 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 140 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 141 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 142 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 143 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 144 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 145 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 146 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 147 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 148 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 149 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 150 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 151 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 152 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 153 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 154 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 155 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 156 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 157 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 158 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 159 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 160 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 161 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 162 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 163 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 164 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 165 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 166 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 167 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 168 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 169 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 170 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 171 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 172 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 173 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 174 ...\n",
      "validation accuracy ... = 0.955\n",
      "epoch 175 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 176 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 177 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 178 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 179 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 180 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 181 ...\n",
      "validation accuracy ... = 0.943\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 182 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 183 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 184 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 185 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 186 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 187 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 188 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 189 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 190 ...\n",
      "validation accuracy ... = 0.952\n",
      "epoch 191 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 192 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 193 ...\n",
      "validation accuracy ... = 0.954\n",
      "epoch 194 ...\n",
      "validation accuracy ... = 0.952\n",
      "epoch 195 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 196 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 197 ...\n",
      "validation accuracy ... = 0.956\n",
      "epoch 198 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 199 ...\n",
      "validation accuracy ... = 0.952\n",
      "epoch 200 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 201 ...\n",
      "validation accuracy ... = 0.953\n",
      "epoch 202 ...\n",
      "validation accuracy ... = 0.921\n",
      "epoch 203 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 204 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 205 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 206 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 207 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 208 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 209 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 210 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 211 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 212 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 213 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 214 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 215 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 216 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 217 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 218 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 219 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 220 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 221 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 222 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 223 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 224 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 225 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 226 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 227 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 228 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 229 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 230 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 231 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 232 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 233 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 234 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 235 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 236 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 237 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 238 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 239 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 240 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 241 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 242 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 243 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 244 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 245 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 246 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 247 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 248 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 249 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 250 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 251 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 252 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 253 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 254 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 255 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 256 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 257 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 258 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 259 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 260 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 261 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 262 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 263 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 264 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 265 ...\n",
      "validation accuracy ... = 0.935\n",
      "epoch 266 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 267 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 268 ...\n",
      "validation accuracy ... = 0.953\n",
      "epoch 269 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 270 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 271 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 272 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 273 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 274 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 275 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 276 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 277 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 278 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 279 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 280 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 281 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 282 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 283 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 284 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 285 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 286 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 287 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 288 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 289 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 290 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 291 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 292 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 293 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 294 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 295 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 296 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 297 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 298 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 299 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 300 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 301 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 302 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 303 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 304 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 305 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 306 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 307 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 308 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 309 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 310 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 311 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 312 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 313 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 314 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 315 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 316 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 317 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 318 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 319 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 320 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 321 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 322 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 323 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 324 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 325 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 326 ...\n",
      "validation accuracy ... = 0.924\n",
      "epoch 327 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 328 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 329 ...\n",
      "validation accuracy ... = 0.935\n",
      "epoch 330 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 331 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 332 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 333 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 334 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 335 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 336 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 337 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 338 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 339 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 340 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 341 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 342 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 343 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 344 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 345 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 346 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 347 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 348 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 349 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 350 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 351 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 352 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 353 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 354 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 355 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 356 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 357 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 358 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 359 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 360 ...\n",
      "validation accuracy ... = 0.945\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 361 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 362 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 363 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 364 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 365 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 366 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 367 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 368 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 369 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 370 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 371 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 372 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 373 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 374 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 375 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 376 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 377 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 378 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 379 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 380 ...\n",
      "validation accuracy ... = 0.873\n",
      "epoch 381 ...\n",
      "validation accuracy ... = 0.931\n",
      "epoch 382 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 383 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 384 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 385 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 386 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 387 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 388 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 389 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 390 ...\n",
      "validation accuracy ... = 0.930\n",
      "epoch 391 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 392 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 393 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 394 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 395 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 396 ...\n",
      "validation accuracy ... = 0.925\n",
      "epoch 397 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 398 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 399 ...\n",
      "validation accuracy ... = 0.935\n",
      "epoch 400 ...\n",
      "validation accuracy ... = 0.941\n",
      "epoch 401 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 402 ...\n",
      "validation accuracy ... = 0.954\n",
      "epoch 403 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 404 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 405 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 406 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 407 ...\n",
      "validation accuracy ... = 0.923\n",
      "epoch 408 ...\n",
      "validation accuracy ... = 0.937\n",
      "epoch 409 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 410 ...\n",
      "validation accuracy ... = 0.935\n",
      "epoch 411 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 412 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 413 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 414 ...\n",
      "validation accuracy ... = 0.932\n",
      "epoch 415 ...\n",
      "validation accuracy ... = 0.936\n",
      "epoch 416 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 417 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 418 ...\n",
      "validation accuracy ... = 0.934\n",
      "epoch 419 ...\n",
      "validation accuracy ... = 0.926\n",
      "epoch 420 ...\n",
      "validation accuracy ... = 0.945\n",
      "epoch 421 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 422 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 423 ...\n",
      "validation accuracy ... = 0.946\n",
      "epoch 424 ...\n",
      "validation accuracy ... = 0.939\n",
      "epoch 425 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 426 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 427 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 428 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 429 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 430 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 431 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 432 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 433 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 434 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 435 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 436 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 437 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 438 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 439 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 440 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 441 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 442 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 443 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 444 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 445 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 446 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 447 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 448 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 449 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 450 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 451 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 452 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 453 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 454 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 455 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 456 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 457 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 458 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 459 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 460 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 461 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 462 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 463 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 464 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 465 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 466 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 467 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 468 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 469 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 470 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 471 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 472 ...\n",
      "validation accuracy ... = 0.951\n",
      "epoch 473 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 474 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 475 ...\n",
      "validation accuracy ... = 0.950\n",
      "epoch 476 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 477 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 478 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 479 ...\n",
      "validation accuracy ... = 0.930\n",
      "epoch 480 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 481 ...\n",
      "validation accuracy ... = 0.949\n",
      "epoch 482 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 483 ...\n",
      "validation accuracy ... = 0.944\n",
      "epoch 484 ...\n",
      "validation accuracy ... = 0.933\n",
      "epoch 485 ...\n",
      "validation accuracy ... = 0.940\n",
      "epoch 486 ...\n",
      "validation accuracy ... = 0.942\n",
      "epoch 487 ...\n",
      "validation accuracy ... = 0.943\n",
      "epoch 488 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 489 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 490 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 491 ...\n",
      "validation accuracy ... = 0.947\n",
      "epoch 492 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 493 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 494 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 495 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 496 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 497 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 498 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 499 ...\n",
      "validation accuracy ... = 0.948\n",
      "epoch 500 ...\n",
      "validation accuracy ... = 0.948\n",
      "model saved into ./model\n",
      "INFO:tensorflow:Restoring parameters from ./model\n",
      "accuracy in validation set: 0.948\n",
      "INFO:tensorflow:Restoring parameters from ./model\n",
      "accuracy in test set:  0.938083927176\n"
     ]
    }
   ],
   "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.\n",
    "# train\n",
    "train(features_train, labels_train)\n",
    "\n",
    "\n",
    "# evaluate the validate set\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, model_file)\n",
    "    acc = evaluate(features_valid, labels_valid)\n",
    "    print(\"accuracy in validation set: {:.3f}\".format(acc))\n",
    "    \n",
    "\n",
    "# test\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, model_file)\n",
    "    acc = evaluate(features_test, labels_test)\n",
    "    print(\"accuracy in test set: \", acc)"
   ]
  },
  {
   "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": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['test_data/03_Speed_limit.jpg', 'test_data/17_NoEntry.jpg', 'test_data/14_Stop.jpg', 'test_data/13_Yield.jpg', 'test_data/02_Speed_limit.jpg']\n",
      "[3, 17, 14, 13, 2]\n"
     ]
    }
   ],
   "source": [
    "### Load the images and plot them here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "from PIL import Image\n",
    "import os\n",
    "\n",
    "file_names = []\n",
    "my_images = []\n",
    "labels_list = []\n",
    "for file_name in os.listdir(\"test_data\"):\n",
    "    if file_name.endswith(\"jpg\"):\n",
    "        file_names.append(os.path.join(\"test_data\",file_name))\n",
    "print (file_names)\n",
    "\n",
    "for file_name in file_names:\n",
    "    my_images.append(np.array(Image.open(file_name).convert(\"RGB\").resize((32,32))))\n",
    "    labels_list.append(int(file_name[10:12]))\n",
    "print(labels_list)\n",
    "                       \n",
    "labels_valid = np.array(labels_list)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predict the Sign Type for Each Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5, 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",
    "# preprocess images\n",
    "#my_features = reshape_raw_images(my_images)\n",
    "#image_list = np.array(test_image_list)\n",
    "features_train = np.array([process_image(image) for image in my_images]) #np.array(process_image(x))#\n",
    "\n",
    "print(features_train.shape)\n",
    "#labels_train   = y_train\n",
    "\n",
    "# Here is a function to predict images and output human readable sign names\n",
    "def predict(features, human_readable=False):\n",
    "    with tf.Session() as sess:\n",
    "        saver.restore(sess, model_file)\n",
    "        results = evaluate(features_train, labels_valid)\n",
    "        return results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./model\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.80000001192092896"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "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",
    "import tensorflow as tf\n",
    "predict(features_train, True)\n"
   ]
  },
  {
   "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": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./model\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAACmCAYAAACr6XxZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJztvXm8JVdZ7/2sqj2c+fTp4XSnOz1kInPCFAkiIKIiKIoo+orjvYjyKl4QJyQqoCLgALyiiILjVVHxMqOvRjQyEwhkIiFk6E56ns487Klq3T/Oya71+1Xv2n26++zekN/388kn9fSqYdVaz1pVdfbzW4/z3psQQgghhBBCiP4hOt8VEEIIIYQQQgiB6ENNCCGEEEIIIfoMfagJIYQQQgghRJ+hDzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCiz9CHmhBrwDn3Oufc357hsT/hnPtkQfm/Oud+/FT7OucWnHMXn8l1zwTn3Huccy84zX3f55z7jvWu09c68p1T7nurc+7q9a7T1wPyn1PuK/85DeQ7p9xXvnOayH9OuW/P/Ocx86HmnNvnnGs45zbTv9/unPPOuT1ncM6/cs799rmqo1gfVvt+eXXQH3XO/aVzbuR814vx3j/Xe//XHcpGvPcPma2/3znnrjOz683sg6v2Bc65DznnDnUYK28yszesV33OJ/KdtcG+s/pvL3bOPeycW3TOfcA5tzE45PfN7DfXqz7nG/nP2pD/ZMh31oZ8B5H/rI1+9p/HzIfaKnvN7IceNZxz15rZ4Pmrjughz/fej5jZE83sBjP7Nd7BrfBYGxOn4qfN7O+8937VTs3s/zez7zvVzt77W81szDn35B7Vr9fId04f8J3Vvzj+qZn9qJltNbMlM3tHsP+HzOxZzrkLel3RHiL/OX3kP4h85/SR7+SR/5w+fes/j7XO+d9m9mOB/eNm9jePGs65G1b/8lAK/u37nHO3dzuxc27P6q8N/8M5t985N+2ce9nqOe90zs045/4o2P8S59x/OudOOudOOOf+zjm3ISh/onPuS865eefce51z/xj+NcE5911u5dfAGefcp1f/GiC64L0/aGb/ambXmJk5525xzr3BOfcpWxmIFzvntruVX5CmnHMPOOdeSqcZWO2PeefcF51z1z9a4Jx7tXPuwdWye5xz30vHOufc251zs865rzjnnh0U3OKc+8lT1XvVty51zv2Umf2wmf3y6l/KPuyc+yXn3P+h/d/unHvbGTbTc83svx81vPdHvffvMLPPFxxzi5l95xle72sC+c5pAb6zer0Pe+8/7r1fMLNfN7MXOudGzcy89zUzu83Mvv0Mr/c1g/zntJD/nAL5zmkh3+mA/Oe06Fv/eax9qH3WVv7yf6VzLjazHzSzdtyt9/7zZnbSzL4tOOZHbOUD73R5ipldtnrut5nZTWb2rWZ2tZn9gHPumav7OTN7o5ltN7MrzWynmb3OzMw5VzGz95vZX5nZRjN7j5m1Hd8590Qz+wtb+QvAJlv56v+Qc666hno+JnHO7TSz55nZl4J//lEz+ykzGzWzh22lvQ/YSt98v5n9TjixmNn3mNl7baVv/t7MPuCcK6+WPWhmTzezcTN7vZn9rcO/uDzFzB4ys81m9loze5/Dn9ML8d7/mZn9nZn97mpYwPNtxYe/w61+6LuVPzT8oHXwW+fcO5xz7+hQNmxmF5nZfadbp1XutZWwga9b5Dtn5DtXm9kdQR0eNLOGmT0u2Ofr3nfM5D+r5fKfM0C+I985G+Q/X9v+81j7UDPLflX7NjP7ipkdpPK/tpWPM1t1pOfYilOeLr/lva957//dzBbN7D3e+2Orf9H4hJk9wczMe/+A9/5m733de3/czN5iZo9+xN1oZiUz+0PvfdN7/z4zuzW4xkvN7E+995/z3ier8b311ePEqfmAc27GzD5pK381+Z2g7K+891/23rfMbJuZfZOZ/cpqP95uZu+2lUntUW7z3v+z975pK/02YKtt771/r/f+kPc+9d7/o5ndb2bfEBx7zMzettqv/2grE8NZ/RLlvT9sZh83sxet/tN3mNkJ7/1tHfb/Ge/9z3Q43aO/6s6vsRrzwbFfb8h3sv3X6jsjZjZL+83aysvBo3w9+46Z/CfcX/6zNuQ72f7ynbUj/8n2/5r1n8fqh9qLzewnLAh7DPhbM3u+WxFd/oCZfWLVIU6Xo8H28insETMz59ykc+4fnHMHnXNzq9d9dKGT7WZ20Pu2RsjMbH+wvdvMfsGthD3OrA7EnavHiVPzAu/9Bu/97tUBuxyUhW273cymvPfhgH3YzHacan/vfWrZX6HMOfdjLgtJnbGVUINwARvu14ft3PRb+w8MtvZfgUNmVv8/WrhXntHg2K835Dunx6l8Z8HMxmi/McMH4tez75jJf04X+U8e+c7pId85NfKf06Ov/ecx96HmvX/YVhYVeZ6Zve8U5QfN7DO2Emr4o3bmHd+NN5qZN7PrvPdjtuJkbrXssJntcM65YP+dwfZ+M3vD6gB89L8h7/171qmuX++EE8ghM9v4aBzyKrsMf3lt94VbEeFeaGaHnHO7zexdZvZyM9vkvd9gZndb1q9m+X7dtXrNM63vo3zAzK5zzl1jZt9lK2ECa8Z7v2grYQyP67YvcaUFYQKPIeQ7j5741L7zZQtCQ9zKUstVM/tqsM9j1XfM5D/ZieU/a0W+8+iJ5Ttngvzn0RP3uf885j7UVnmJmX3Lauecir8xs182s2ttRSu2Hozayhf7jHNuh5n9UlD2GTNLzOzlzrmSc+57DH9GfpeZvcw59xS3wrBz7jtpkIkzwHu/38w+bWZvdM4NuJVFWl5iOAE8yTn3wtWY6FfaStjpZ81s2FYmk+NmZs65/2Gr4t2ASTP7X865snPuRbYy0P9ljdU8amaQW8SvCFv/2VbCdG/13j+yxnOG/ItlYbhmZuacG7CVScrMrLpqhzzTVsTKj1nkO2aW952/s5UIhaev6gB+08ze9+hfbt2KrvZJZnbzWVzz6wL5j5nJf84I+Y6ZyXfOGPmPmfWx/zwmP9S89w96779QsMv7bSW88P0FH3Nny+ttZcnUWTP7qAW/7nnvG2b2QlsZKDO28mvbR2xlYNhq3V9qZn9kZtNm9oCthHKKc8MPmdkeW/mLz/vN7LXe+3AwftBWRKvTtvKr6wtXY6/vMbM/sJUP7aO28qH/KTr352xlsZkTtpJ77Pu99yfXWL8/N7OrVsMMPhD8+1+vXrPwV2Dn3Dudc+8s2OXPzOyH6S9gy7byhwWzFW1nO4TCOXeDmS36lWX6H+vIdwLf8d5/2cxeZisPvWO28geqUCfw3WZ2i/d+rX9d/XpF/iP/OVPkO/Kds0H+06f+4zBsVDyKc+5BM/tp7/1/nO+6mJk55z5nZu/03v/l+a6L6E+cc7ts5SNqm/d+7izP9fdm9k/e+w+cxr7/x8z+3Hu/1r+QiT7hPPrO58zsJd77u8/mmuL8Iv8RZ4p8R5wNjwX/0YfaKXDOfZ+ZvdnMHrcqmjwfdXimrayMc8JW8jm808wu9mtb2EQ8RliNGX+LmY157//n+a6P+NpBviPOBvmPOFPkO+JseKz4T6n7Lo8tnHO3mNlVZvaj5+sjbZXLzeyfbGWVyAdt5adifaSJHKvx00dtZSWl7zjP1RFfQ8h3xNkg/xFninxHnA2PJf/RL2pCCCGEEEII0Wc8JhcTEUIIIYQQQoh+pqehj9//+v+An+9wUbliivaNcj8Krl/EYlgP/jVyLffDpHTo2ZzrbI5dy7n4/t/zmmeduwsT111/JVwsDa7N3c/14vI4jtvbA2UcAi5F32k2m2A3WklmRPh3jpQuFF7HzMxFQfNwHfmXbbLToF5JktCuuG9E9XJBPeJyGcqqZPNYSoP799SSSdLCYx2Nh/A81DgtqvOXv/zguvmOmdmFF1xw2qEDvCOPgZHAZyaHMEvBaIT7lum+L9tzaXv7uqdBBgTbe+wY2Hc/8CDYh08ebW8fn6MFsyoVMCc2bwP7wh0XtrevvOwyKLvkwh1gJ/N47ql997W3F/YfgbIjB46CnbbQJ47XltrbjywuQdl0E/dtUVslMMaxXXlaOnjw0Lr5z4ue+xyomA/8vkpjPOY5IanjyVw2jgcHBqGonOJc8+V5bNvDl2S+82M//JNQ9r2XPBHsrYObwU6S7Nz3NZeh7Nf/Ften+uxf/A7Ye0pZztdyO0PHCtUS9UsJ26NcyuaXckTzocd5rEm+0wrmubTLfBlRP5Qr2XVLZRwbLsI5778//pl1nXue9awnQ2VbrczcfxDH/NIy+ounMeECm1/eyvQsS2mQ4PjCstwzM/dMyfYvl9DHN28cBnt4DNu3smm8vf3w3oegbOlkA+vcQv8KZ+MoxmdzpYz+UyFfTKpZe2zcNgllkzHuG6dYj0NLmT01jeMlmcc+evjA1Lr5z/DgCM491tni535RsFyuwvQPawu06/IuE5w8f91iP1wL/Jz2Bc+Ps8EV9sLaWF5eOK2K6Rc1IYQQQgghhOgz9KEmhBBCCCGEEH2GPtSEEEIIIYQQos/oqUYtH0PrO5alpBVyrnMkqO8WcMv1OEcLXZYoXj/GkGmLSria/lyaxfoPUiUih1oXhjVsRaxFo+ZJJ5A/1hWU9Y5cpgQflhXHDOfrne3RSklnxRcmv4xLYdwzacFyYjmsc9oKdGbk3zktgqNzB/dQNI7MzCI+NtyXdGVN6n++f2g51s0lqKnxUWff8TxIXW//RpTXk2bXj6jeCd8n1bVSHWlvjw+QPsOTP1VQOzOybWN7e6o5BWWff/ALYC/XUdNVHsrmm8u3PwHKJi9ADcbERpxPhocznc74CNbZSDtUGdsI9sSWXe3t5vQM7juIeo2khv404jK9yjjpT+qktVzidgf/YR2E9Qz2j1Kc3Qdr0nxK44n0UeHcU2P9K2nUBj0+nutfuqO9/ae3vwLKPjxwHdjXX40atuNTmbbwq4e+CGV7j90FdrUxAXYSZfUox9TwNLcW6WMTFvEmND9S24XzZ8RzOOvd+LowX5KOMO7t3DO1gBqnxYWF9vb8Epa5lOdurHuoE05Yo1TcNdAmrBnuIpG28HHFz665JdR3LbRI530y0LWmOC+VqS8a1DXhPF3FprDdE6jxvGJiBOzGhux9664TC1D28HHMi1ymh/dyoCNMaljWbPXwPYg7sWDiK9KG8aH503abUIvWKqA9Xef3k+IadlsTIfcvHa9zGgcX1mRN77q5d4ugiK9yhu/Q+kVNCCGEEEIIIfoMfagJIYQQQgghRJ/R09BH/tmvcMn9iL8hOy+5v9Zl8s9VCF8rwrCfl30jLmv9tr++GewHH87CHK587k9AWbSGjAJrrX/R/vwz9Sn2OGf1OBs4VAPC6rjtciEfnc+bcPhNLkKPQntcEH/B4SJUjzRXsewAXuKVl5/m+w2HQy4KqMu4CkO3eKykvNS/8f2GS+vSdXLhMRTOGfaRo1Clc7hc7unAV4NupfBPnmviCobshWGEI2UcP6Uanqk8NI7lE5k9lWI4TnlyC9gVCrG5YGJre3v79l1QNjKKYT8lWro6jDCKKcw68dxvyMbJLKyyOYV1nDmO4d1Nh2FQY0nW73V63DRoSewGLd/fCnyGw9d4vKwnSR3D08Jlz3NzID+LInrEwv783MJ+GBpA37p6aEN7e6mGIbbTUw+AffPH3gj2cpgKweP9bCDbDH0nTbI6J/S8oOkzl94kDGekyDXzPE85bqtwef7Oc+mpzHBejyjcu8eRjzY9h+8JLskqUCnhUvatRpeUJ7nw8gxuz6Jwxm7Dh8N9w+cVP5sWFtF/Si2cX8bHsxDF4Sr2YzKI12mSlzSCObC1jPPD+BDOeVfuwJQki+NZ2z5C8/LBvSewzuTXYRhp0iQZTs6T149cN0Fc3ZnPgWteBh8u2+XYs1liP3eq4J2pi3+f4ujg2NyJiytSeN3OaYjMzFxBjOmZph/QL2pCCCGEEEII0WfoQ00IIYQQQggh+gx9qAkhhBBCCCFEn9Hj5fnRXlu85jnUSoUhpIY6iXKKS74OllB39oRtWQz5tuoilM3egbHoP3DpJWCP3pAte12r7oOyE1WM6z5yEmPXH1za0d6uV1DbYh41NKWCWN1uUriipeyZfFqE9cORximMk2e9imO/ysXbB+fJBfJjC7H+Kw5i2dcaqx2eqURCCb5OktAyzcE9RI5TV1ihHd4+S/1yt8/aufC8uGvOM3JLOhecNy8s7C1hfSJa13qEltQfHcSxOBxly6gv1WhJdYdL32+9YDceGywZXa+jvmvLOOrZxoa3g71tYzZ/jIziPFUuYZ2jCOschTH3PHBpWfSY/Ksc2Ht2X4SHLuJctPcrqJVqLWW6kg0lfNxE46NgJ4b6nGPLmd1iZ+uhPjaiNBQgE8nprLgfSKMQZ20QkT4pruBS/mUuDzVb86jXiWbxOVWp4LPJBf7B2lLW4LTwdmlg4/10fZ6EmjXuM05JUrRENi8nz1WkSifNUA9Ml00a1ktiGm8XTWbj+vjh/VA2yyldSLcXB/6UUGqZlASDnvSC+MCmtub0BuS3cdDTrGtPaV9ecn+knN3DUJnTAdHzhgb6YiO77kkq2z+NPv6Jg8fB9iezY9Mm6tkswXFab7COsbOum1NynC/OZo0A1qLnl6en/dfyrl6g2+2ywn7+PaHDeU51HU4BRKVdji1Ke0CHdq5h7ty582p5fiGEEEIIIYT4+kAfakIIIYQQQgjRZ+hDTQghhBBCCCH6jJ5q1JhzlYcryukXSGPRwtscTmba27t3HIWyLX/xfrAP3XM/2MeCmNPjVP9cHinOWVWQm6FEMbQlKr8+uMnht9wEZQcPYLz1w9WLwS5bpiOpR0NYJ/pWX0ufsLZnPcnFxfswxw7nI8Jj+a8RoYYtl62PtBAp68GCWP18nhqOiu6cCy0X507alhbljcKKUt4vz3XkehXkE+E8enRTocn+zSLFIm/IxWr3MA/Wqa8faA1pGiyl2CYDDdItJllSnmhoDMo27UAN1/jkDrCX6pmWxpG2dNsYnmt4CDVc5TjQp9QxMRDLblwuv1XmM75F81QL9RlJA889E+QRc8uYM2lpAfdlfayrZVoqnlvGB3AumhzdDPZiMpfVYQmvm/RQo1aNOs89eZ0mtiUP1XI5mHtIs+dJ8VWrY6fOLGfzeKOBmizWKI4Mo94tiTLbkyatvoTa6kXyh3B6iWi+iFn/y/Nn0EKe8/WxRo0lbMF8w/Mya5tYC9hsZQ1fr6GPRuXe5cEyMxsdRP350FDm95UB7KeYBnLumRLMYxWaumPShtUa6HzhsI953mfNDnn25sBXN5B2cpGc3JfQfwZC7WkTj23U0Y99E+1WPcijRtrJI9PYr8dmcI6IgjERl1DTyXo+fpcJmyMnuSrIZXeu6ZIRuPjg3DO2aJ0HPrTz8zl3li56r8L3As4lWbgv/0Ov8sh12Tc3dqCQL7yG62boFzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCiz+ipRo01CmuLEz19OA/Q5UMPg33gza9rb9cxrNkeyWl00Ib0Vt3SaNEOgSymMB7fzCwijUkShIEf/8XfgrJNy9iNz/6dXwH7k80Lg+tyronTj7deS461c02oCzEzS8LcT5xHjHPCUDx+qJXgfDGswcjpKsJ79rQv/92D2zoo5rxpOX8gOwk0GTltR64b6J5CvRudN+Y/1ZC/t1qhFpC0cByrn0usF+bu4ir2MAnfygXBCvPIsEatUsJcaKNV1FVUAx1JlTqqQpqk5tQ8Xnc50DelOPmknrRhHvMCJUmm9UhIR9QkbUfSSjraKR3bSmhfOpeF52KhCOevqmNuo8iF5djOTdLVtSLMDReXMu2UN9KYsF5yHRmkMbLczNrPR3hPSS6FI7ZPHDR9Ke6sBzUzq5M20jWyHZIa68hyajk6d9aHlSrqolqkhWP9Vzh4eT7kiYrnrVBjxc+POKacc/mJrGOdotwcTwcEfZR7tlrZeolroQbwSJB7cLbEWmS8rzLNqY3AwYarqPEcoDaaYo18MIwjelbxaBocwDZ62mSWw/FHvuuFUHZ8HHOU/ccH/zfYdxzIcsXNexwv5QHKBzmCc2+zFcyJ/G5GPt/iARSI8lIahyXyAdb3WRTM06Qfrw5TTrYe4gvWOcjn+yp4xubGadeMiNm+BXU6lR0e2+2xv35vlPxOtJY8aqy566arK8hXeIavPfpFTQghhBBCCCH6DH2oCSGEEEIIIUSf0bfL80e5H+M7M2AYbvOMgUfAPvy6t4M9FoQPLOaWdqefRPkn0iCs0lHoW5zgz+nzFD4Q+SzEZHSZlkguY/hR3MKf21OXlbsm1ukkLYd79A0YGvm0n395e/uTlcuhrBlvRDsXuhOG+uXWqbVewWEyYct2+7k8lyYhNCjsi/d1/DN3EL8YUZ1cl9DHMISoW0oBR0sex8EOnsN+8NBcyENYZ16VOU3Rd9KC9ogoPCQfJto5NCXNhWL1dnn+3LgOyygVwtAQhuMMUnhSaSkbx9EihuTVDuLck5w4iceWg/DFBI/1KYf30lLvsHNxagT2rzAkj/uCr5JwX4VzAIX6sZ0OYjhWuZLZCc0Xi40S2TgnNsKQy1zIdu+oU+haqZzdR8L9QL7UaFFYYRAq6ihE39P4ShrYD5Uka69GitdZ5lBYTyGpQR86muSr1N+VEVxOvhQsx56ba9jPyIfDkEUOSbeoOBSyFIfL83OYOZILbwyuxfvmlqZfZ+YX58B2wdLwJU6rMILjp0ZpKYKITmsl2E8JvUOYQ9+rlLL2HqC2L1XwXWViI9bjeT/0/Pb2U3/kJ7FOFUwzcvFV28D++GdvaW8/sH8/lB07MQv2zJEZsOvB7NSokH/QfFKm8VQK0jA0Wng/Y0P43hPH2M6z9azPaKjZJZc9znrFWkLyTpGYp+DMnd8RVk9+2mfKnTk3VwfvH/m1/cnk96/OIZf5pe5P/x2j6+vHOi39f6bvPfpFTQghhBBCCCH6DH2oCSGEEEIIIUSfoQ81IYQQQgghhOgzeqpR4xD1onhN1tkkEcbcjwTLQL9oFJex/tiv/35hPUCjQSGyk7Rc/5eecinYj/++F2V1SjgeH082zrG6gQahXMaY8KSJGoOTt9+J5R98f3vbN6nbPKlMGnjd+978J+3t7/qfPwZl/7LjWrDT8gY6dxB/nosfpuDtdYR9pxJoQdK0sx7HzMxo6dlQz8BLzLPfsQ4rjkvBvqzl6bwE/KP/0j62i+YmTUg7loZ6jeLlYfncUdR5mXzWdrAdNh0viZ3bt2j5ZNYJ9lijxsHxqL2j9qJl01sJrSMfLHUekQ8s1VBz4Vqon3WB7oaXHOcloodINzJWzTQpgwO0NHcVdSJ8T6HurEH9tlTHtADLDazzss80WlPLOA/XaLDVW3jdZrBEdrhtZlZrUds10OdrzaBerGfq5d8Yq6gFKgdaH2viPZWHcHnxpE5+lwTtx2vZ89LsjufXbH+6rC1Se3B6Bgv0X4NlfH6EWjAzs9Fh1GiGesiE5p7lBur3apw2IZg/WYOW16SR9jaYfBKaD5t0fyyfDv2f9Tdx1LvnlpnZ1j3bwfbLQXoPR8vV0zx1ZAl1rGGbLDTxZWU5xbEZOZwTxiqXtbefdeOToezFP/ItYG/fswXsTZuyJelT0oZFVfSXx934TWDvenz2jsH9ljSwL2YPnAD7X//hn9rb/3b7F6GsaTgvV8lH6sF8cnIK/XLzIF635Wi8DGb9kpKO0MXo873lXKlzi99V8pmYQNVdWCMeb+FS+N3TQZ3+NwG/zOTSYQVzUT6lAF91LZq84rQIfOZzgX5RE0IIIYQQQog+Qx9qQgghhBBCCNFn6ENNCCGEEEIIIfqMvs2jlot7bWL89bPL+9rbn3rNH+POFIKe084E9vTmTVC27ed/FuwnUUx1M4x7pfO2KJcNx+CHeZJY38PaoY3XXA327A3XtLePvxHvd3wWdTGe8omEsb23vetvoOyFb/k1sN8zhxo1Xy6IKe6hzIj1PGHTs1Ys4vjjoiBizzor0mG5ztoY1gEZ5b3ilosCzQnrJEiCYxXWaAa+k+TqXKwzS4NylsWkuWMLcpGQHimXm4w1a9ZPcEx+esptM7NGgnPNIo3rwVBLRLl9aqST8GXSDgU+k5ImqTqM+qYLR9GGiS1G/Vorxqk8oXtoBv6T107yVSjPVpBXbmkJ55pjC/NgLzRwvqwHOaNarE/J5eBizWe4TboIHjDrSIs1F2G9SFfFU0KF56ZA75JQgibOFcZ/Rw3n8SppnEeo7cbKqBsaqgaaxhrlduN+mMU+DfstiVhnxro6mrcCs0THlkroszH3aYEuhufpJudRC/NOprkHgvWSJY/tnQSazxbdV5XEduUh1JlZkHvROxzjw4Oopbx2ZCfYz37qy9rb3/nS50DZlp2ou4qpPadvvbW9ve/IMSibq6GmtZFgLrTQRXbtvgzKLnj8U8CeuGwH2D/+8kzfd+W/fRDKPnnbJ8D+yjGs14kDR9vbLXp2LZPObNMWnGsHg/3rdGxtDq+znvC7TUgu52suP2A+w2q2VZyHM5ehzXU+luFhHJr8rtLNxqNZG+c772rcHl3WBOB6WOe5Zy1vNrlvHOVRE0IIIYQQQoivD/ShJoQQQgghhBB9hj7UhBBCCCGEEKLP6KlGLZc7KojXzOldkgrY3zz+ANh3vuYd2bER527hXFhoX/XaX25vz5Qw1r+ZcH6WzrmhErqsp+/e1HNcbKBPobQdjoVVVOfRWlY+9irU0W2axpjwB9+GGrbw/h2JKG575W+Bfe3bXgf2HUtXBVUibQPXeR1pNln/FQpFcN9uYcBhl+bzadC+1MetUM9EMeHoSWYxOUgc9EOc843iY6FipPXwdC5uqzBVTZ30GrnY7FzYd3ZuzinH+r5crrLwOt3ytKwzrHEM49s5R9PiMuYnymlRA+1pRLnB6k3UPniH515uZOcaHh6Bso1VnI7nWniu+emFoBKcEK/YB0I9WLmE/VghJ4+pXwe2Tba3R0dxXp5pnMRqNDHvU6iNa3E7dsnhV6RPyA3MdaQ8iP2UJlm/xNSW/JfPoQpwc4RsAAAgAElEQVRqxULNXpO6sEW+5DxpusqZjmbQoR6pRHolR77TnJpqb9cW0L9jzsvIutxA/xjRG0OZdGYs0qsHp05pTvMxPWsj1uQFZeQbCWnWebp0rnO+S9aDrzctj/cZVbP73LVnD5QN0Nx++MA+sEsjWblPsK13b34G2C9+wreB/d0vvq69XW/dDWUH34d6rw2lbWDXmxvb25fsvhjKWk3UNPoU3xNKpcxXpw9h2fKnPwP2sYUHwb7g+m9sb3/bc14AZXFjF9gP3vxRPPfS4fb2UhOvuzSDWtv5FOe1SpA7s1nHsVTv5WtzLv9nsN310IIcsV20U6xDQ40aU6BrN7OUX3bxxGSf/m9HuXeVwrUJit8/+BshnMm7tXNeK1jwcnOGLz76RU0IIYQQQggh+gx9qAkhhBBCCCFEn3FeQx/DnxsT+snzGZX7wD5y05vAroThOfRrYo3CL3b89qvAXgy+T0v0reoNf+ZODH8SbwYhFxz2wffHy/WXghASDvP53Te9GetBv7defXW2XP8P/uAPQtnURlxSv/UNN+DBn7s9q6Nhneq0vHL0a78HdvU1WWhkK8JwiF7CS3uHS6+uNZQl/Kna80/trvOS2LlrccgQX4jqjGGG2A85CpabLsX895VuDRCEy1BjtTgCoiAEILekbatL+FnBkr6ux38jyoc6BNencOeFWg3sBtV9IYy69Ry+ReF8hiFacTULA7qIxu1G6teZE7QMdNBZgxSONzQ8Bvb4AC43HS6jnjbx/pIlDF1anpkDe/pEFiY0tnkzlO3YsBHsiNrq+Gx2rgVKX5FGHNbD6R6C7fOY7IF9tRWEmTZpjNco1UPhsRSvR6uAW+wwDG4gDEPmuWcJn1tLi9N43SBtQlTFsMnK4BDYg6PoW1ElWyJ+kMITa8u4NPvJGoa+1hpZn7donHFEVIXmpnLBEvv8/Myv6h3MPY4lCb0LmzUzq1LdN45m7T1/4giUHaY2GqB4+isu3NLevmxyEsqecOl3gf2Nz34u2LXGx9vbU3d9HMoiCtxfHh0He9P139LerlyMKQPwDamYiX0499QPfwHs8pG9YB+765Pt7S1XPx3Kbvz27wa7NoW+9+eHs5QCX6Q5rrWEPtCq4z2VAl9sNDFUuMH5n9aRnIwDHmTFS+zn3kfgedxlaXtODRJGTfL4yVWS5qYgXjqi+YNlHBE/E8JtnvNyIZb0fAnHEj+nc6GQBWHXVjxfFIY6dkllcLroFzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCizzivGrUkWO59IMHlU3dEC2Afp1jP3FLhAX7XdrDHbBTrEYTop7y0aA2b5LW/8Ztgx8F1f/03bupYBzOzUhnjvktxFgf92te+Fq9Lcb8J6Rfuuuuu9vY999wDZTfd9BqwJ17wHLCXbsviwBskhKCQYGs1MB77yjiLKb8Ld+0Sm3tuyaVvCJd9TnldVj66c/Q2x0THuZhp9LPQO6rkg0kD9UisYQuvm9OC0XWimIZmcMNJLkacTkX1KgV6jlIJ7y+J0W5RPcJV4Jt0f0luSV8rsIu1f+tNPow+axPupoTqVqeq1mF/jpvHvqmSjnXz5on29kgZ237p6Amw0yZeeNdll7e3t++6CMo2btkC9sQEakwGgms1FlGDtjyDeqbDDz8M9h13ZiN/cQaX4x8ZRW3chRNom8/mj2QWdVS0inc+vYnD0hAet+tJbRm1D81A7hCVUKHjaf6IYywPdc0ljw3QbFH6E07BUMvm5sYiLi9eW8AULdw8Gye3treHJ1DbNDGJusMR0k5acA/lGtbx5EHUFC0eOQT2fKDpbvC8ldL8mbC2JbwJbNdw+X0zMxpKqG2hwd8tfcu5Zu44vsvYYtaGrBm3EdQPbqDxNDqY9cU3Pv5qKPvW51wDdmP6DrCP3/np7DwTqC0d2fU0sN22i8leixKtM9XdqAWrbroW7LFLsF5z932svX3syzdD2eRV+F73vBfdCPa9R3e3t+/55CNYEdYlL2A/1HxmD2+gfZd7p3Hk957wWdVlhf38W0+YDiunD85dGM2COvKhEb1TxMEc6SJe1+H09fY+4vcNsmk+sSBlSUopeFjvlnuOh5o8nra6qwE7wmmCThf9oiaEEEIIIYQQfYY+1IQQQgghhBCiz9CHmhBCCCGEEEL0GT3VqHGwZznNLn/dFox93/vKd4BdYi1ZYI+mGBM7+XM/gdehPAhJEJ9botj319z0W2CXSGPyW7/9G+1tx3kdWEdlGNt8069lujSOm7/pJtS78XVvu+229vZHPvIRKEso3nqQvr+PXvW49nb17vvxWEP4uhuTLL4+Ib1FiXJErSu5mOkgJwjn5uBDczm8Aq0Y7cvalzKduhz0ecSaNGrMQcplNTSc2VXKXVSqoJ6xTHazno2PBuX5ai5j/pg658Vazuyyxz6MHWoiOJ9hM4jtrjVwjHJbsS4mbGfuoTON1T5TUs65EmrUKAY9l1OFQ9/D+HXyyyHqtx3jqDHZFIyv2aNHoSypYRz9Fdc+AexLr8w0KCOUz6w8grmvyoOoBYkDLUypROOB8rdtG0eN0mIwvh68624sm0V92/gG9PntgV1vokat1cTxk7IeI9hm7Wgc906jxvkywzHiSIdslPssrgyCXR3I+iWmnJ1GOchai2g3g/L6Ao7xhLQfmy+4EOyJbVkOzOGNqAOqjJCGm+7XAj1YRK8MlSGcx8okFisFvsNaj9ysEHXOT8VzfKVc/OoSjveE9Sk9nnt8Ha8/F4yD0RGcjydpDIxRm9UCgexFT0ZdWb2O+tGZ2z8D9mglG4tDl3wDlEV7rsMLkd+eO0kxnShG/7HJS8AcGc6edWmC93PiTszBNvnE68H+zpf8ZHt73zjOS8f27Qf74EHUES5VsnruvBLn2vmU2mYd4fdEeJfhsq5rBvhTbp7qWH4+h9prrlNek1bqXM51zAntyAyvRXXiXIo51wr1+FSWNmlOIP1smA80V+VuSyIE/5LrkjMUyOoXNSGEEEIIIYToM/ShJoQQQgghhBB9hj7UhBBCCCGEEKLPOK951KppFn88YvugbNqhfqHMqaOCUM+7t2P88Y0V0lJxzGyaxacen8b8M66Mmp35mSNgDwSSBGeoT2CtWNN3zhHx1Kdizo9h0pSYwxjab3jKE9vbH/4watTuvOt2sG944g1gX/LDL25vH3z166GsRVXkeP5PvOVP2tvlV/8ZVrGH3/mhvsuM4oR9cQavovJmA3UiJNexEukbIR9HgucdH9+E9oYJsAcDPUdUoXwilN8spusODmc6kqSJftaoo2attoy6oWg62782j/u2GhSrTfqlMG/gQBnHVUpx3RznDbjOGtOekMvDl91XRHondpecjtGF8evUXoZtVGlgPzcOZ/PNEmkcL7/+KrC37dwK9mIty5118uAiVmoY86aNjqHvRY2sbw6SPuORwzjHbbkANUzXXJbVKyZd7t233Qp2bZrm7cDPh2Kc4yoR+kud3CcKfIZlqDl9wjrSamJuSRdn876nccq54HLpEoNnUeRxHOfuMSHt6WLmO55yjo1vQ03a8KZdWK9A03Vy9hiUzZNWcmwC85CObsh8q9LC+WKR+mGBdIeYCo5zKSIx56wMRK+UQolTsOX8AfIksR6nx3PPrj07wZ5fyLRkQwP4DlGh/JkppVmrDGV6qc0XYg6y2udvAXtuCjVrY1d/c3u7tA31XPVlfO9pnsD5ZfiC7NnlqsVaKFJeWpi1MZrDG5q/+wGwd193BdjRaDb3bLgU2+rEQx8F+xDd/64X/nB7++lPOAxln5n7V7A5J6ofyXRoKek/azxRrStnIw7kZ25nbX7uDYme5eH44tyyrEnjk7Va2ZyQy5vGevACnV2a4nyZy42Wy7UZ6GOpD+MY9221WMcKZ6LrFOexxXvITT5nhH5RE0IIIYQQQog+Qx9qQgghhBBCCNFnnNfQx4XSlvb2+KE7oGyef12kn1vDZTu/6Zd/AXfmn2KJ8CfRX/+134SyFi3X/653Y5qANAhvdBSeWOLVhh2G+oT3v2fPHiijqDIzCqtMm+F1aQn5Mi/tzsu2ZseW+KfnLiFE1eUg7CfC+21w/Mk6QhFXloQ/p1OsWkRLZPPP7eGv+nwHJToXnztcJnx0GJe13jCB4Wa85H7TZ+3nKXwxcrjvAC1zHS7Jv7CIwSU1Cn10tA728FhQzyY25HID941T9unOy9S2eMl7LAY7v8xwb4l4+eEgfK3bErqcdiCsfZpbExhDiOJB9Im4lKU4iGMMqds4OQn2sYMPgj1/Mgt9GxnHJaPHJzHUbe7oLNgzM6H/YB+PD2CI5fIshj0dqmYhVJsvugjKxh7B0KXW0eNgl0tZOw+WqG0iDJOLHYW2BE3LYS/sT+tJi9IKuHBMUHh7REvMJy0cm8uL2T3z/cYUhr20iP1Qr2W+Ux3AtA9jGzHsOqpi+xydOtHePnIMQ11bCV63dOIg2DsvvTK4AUzRMXPiBNgnlvBcEGJMA4kzLPDi/GHoK/d3krJNB7vwujzL9zJ0zWxoE4YlD4xnY6I+j+21XMf2bZHfP/3aTNZQiXCZ+NYh7IsSpYCp7NmdGWVM5/HAZ+4Fe9f2LWC77cG56JlhFazjguFS9w8E7zK7BnAOmCCpSe0AzluDFwbXreLzNq7idUcW0YOqwdL/123aDWW3N/DYAZYTHMvm2iPHsV2PL3FSo/6g25wI7425cD06F78XBufOpa8hO/UUr2vhOCaZAfk3p+GAKlM4Zr7OHM6Y7c8pVqISvm+5pPPzJVenXCgkhZgWSUDOEP2iJoQQQgghhBB9hj7UhBBCCCGEEKLP0IeaEEIIIYQQQvQZ51Wj1hzJYpm/8tb3QNk464poKeNwyf3mEMabVmktX16ev1TN4rOrFYzVtjrqJn7l1W8AO9T//N6bcKl7XlI/Io1SCGvUKqRJKow3Jg3W8ABq4WJq5zBGf4jiZ3kpXb7umM/abqE5j9UooU5iXSmImeY6s54tp0kIymPWTnHsNsVUl0tZWw8Po+9Uq9gPLUOfDZeA5cwNLKtcWMKluZcXM+3C8jL6WbPFWh/UxQwMZf4SVylW2+OxRjHjYVuxVIt1X9zuYb/kluNfhzjuInhMhKkFUqo4LxOel7AF95Vbfh3/oTRAyptg/qBdbWILjqe5BmojkqBvrt2OS37HFdSjLLVQ67JhY9bv07S89jVPwqW679v7RbAPHL+vvT0yeTWUDYyhbqR5DDVqlUCXNEC5L2LHusXOS6zzGO/lAutps/MY8TG288AQzgmtJo7VcBxXKCXHMD3zGqQHS9Ns/5ERTKEwPIE6qKVkCq/byPRuOyYvg7JtE3h/x5ceAnvu+P3t7XnKClGjOg6MoD+0atkThh9p/D6Q8Fwc9H/awnmpTs9pFqyEKTd8Ll2L9ZS9+1DHuWEwE6QvzNI8yFpl0ho+9Rnf3N6uLGLZ1EnUhsVXoZ40vjhI4UACwZ1XXwz20ATWY2Ema+/lFr4HbNlO7x/pAbDHXaZDG6yiVmzkGpzHTjw0B/axRzK95O4rUO87vBN1mY1bUVuZPpi1T3UXtsWRBXxGnqRFERpL2f3OzNHaCguswfraAObQnMCL09cUvX/S85GflynOaxG8j+O+oY5s9cp4rqBeOZ05Lbkf0Xt/M5h7c/fDa15w+p5g7uF3woj1srm2KlLgn9nko1/UhBBCCCGEEKLP0IeaEEIIIYQQQvQZ+lATQgghhBBCiD6jxxo1jM8cWM5inUuUfyWlWM4SxX2G8etxFWNiY9KccCz8S1/6isyg3C5Dw3iu5doM2FEly13yil98NZS984/eDHajgfH7Yb6JsXHUlJRJo5aSJqtczuLAI2rHoSHKl0I6pPBc3OFpl9wby1F2bGIYi57PT7OOUHuE3Za7A9I/5eKTA4FYHGOLsO9EdPZynPVThTSIEcX9tygnTjMJcmhxf5M0rFbDY2uNLGY+ijHp3gD1Q0J51ZIgIZWvYB96TgnTIJ0liNToOjkxYGdfymmMeqxR45qBRi0nn2Nfo5jzwM4XUS6wBupGkvlMO3Ttk66Bsk1bN4C9sIB6pxMHjrW3/+Wuu6CsQtqozZOoWRoKci0mHn2+mWLuoo0T2CAT24L8bkOkRyAJXitB/3GBfpKVHSlp0lJyRpgDc8nuepeJr0S+G47rgUHSSSRN2hdFXaHWulrFPnKkZ0vJDttgaGwYiuJBzEe1YXAb2BNbs+fWaIRayNn9d4O9QDn4kiBfVUy5u4bGsR5D5FsLgVauTveT0xYbEc5rNEibjWKNWjl4BrqI9Td8ofVlfhbnY78Q1D3F9kqG6Lk/gGNicCh7H3HTd0LZ2BiOzUXKr2phnr5h7MexC9F/mFIpq1e1WS3Y02yQ2veCYK6tcq5ZGj9uA97v7MFAsxZT7sgtaNsG1MYdeODf2tu161CDt3ca3+sOHUK7sRA8f+kZmfQwfyw/uQpdN/8ihGaYl5L1oCzE4vkVnnnFedRy74VB//NaEzzqnSN/COqRH7fcD6RTDe8/d3ssLufvi05nzX9PFM1jufyxZ/jY0i9qQgghhBBCCNFn6ENNCCGEEEIIIfoMfagJIYQQQgghRJ9xXvOolUtZDHU+lpPzPHSOzi1RDoiI4k1Z7+WD/d/+tjdhnaLiAPb3vu9D7e3/+K9P0Hnx2MFBjAMP72lgAGPCY063VBDMymUjI6hPGR9H7cPCQqaTqeZyF9GFiXqQByopoxbO9TAhTZpgbLMDjVpxPYpcKY5I30gx1DH5Urmc/W2D83YkKerKlmqYb6aVZtqEqqOca3TdxGO9kkAMNFDGY8v095YG5dAK49FdTreAptE9gckuycK6nG+F27ksbNZLEp4DoD5d/LgosJwO5SaJOKeMz/SF+/ZinrSdV+F0PLAZc/9c8czLs8tWUM82U0Mt1NwM6jVmp062t0cinB9qTazj0gL2zZe+9NX29uVPxPxdKc0fEWk+w3Z2NMlF7It1bDzQobIuoIcaNdZcDFQrQRlpXGP2a3SQpaWsnxyNgQFqy3qCPhvm92ONa1xC3WqVkvSVg1MdPLIPyg7NYO6q6iT63USgZzp5En22XlsGe3gU/bIR1DMh/WI+/xACaZ9onh6gZ2vu2FBYRBoa1kKuN2kDr78czOWOcq8OkGbnxidfAfbGIOfd8v33QVnSQm3hpgu2Y0UGCnRouWck9s3D+7N+3/cI5oX79mffgAdHmEuv3spOzs9Tlio3Z+k5X+cHVEZ1CPWRc82TYA8PZzkda4Y511rLuCZCYwHtVivrI9Zz+X79fSOX8HO9LsR6c353Z41aVjF+F2edcpy7hyCXJovJ6dWV3wOLXl34e6K46brlmGORO1/s7OlTjxNCCCGEEEKIxy76UBNCCCGEEEKIPqOnoY9M3Mx+ij82jqEbe6aXTvs8lQaGX/gqhkV89atfNdwhCx8YGMDrtpr4EziHvbzwhS9sb//Hf30KT+uLl9gP4dBHZ3jdmMKEaq0wlAPDA4aHMQRgZgpDAMZPZKEtCx7DjZIuoZ5Pe/lL2tv/TuEnJde7EJKUYsrCaFcOBeXlmHOhj8FP2S0KL4roZ+1KCfs/TKPgPYatzC2gH84uYUiRD9Z8bUW0FLlDf+Al1NOg35IGHlumkMNSGf0/DRZGL9P9lGJKIUChOGFoQq4PuoQrh93AoY/e9/hvRLyELsRVcb1PP9SB7yMhf6o3KSWJy/pmaZ7Ou4zja3x0EssHspCio/PYb0enjmA9mlg+XM3CZWencbnwux84iPvSUu+jg7va24M1rGNrHkMu+f6bwby1QOlKOAVFLpwxmG84zDrq4d8YSyWcM8vVsH2wHp7myJx/tLL7aNA4dpzegysSppTIpcZA0haWz85koWvHZw5D2UKE9zc4iH43Pp6V1+amoCyuk3+XMCy7ErRdzeEzjuPe0pwdhHrSc7hCKXk4pUYrCcYoBzLlliJfXwYoPLYVzCctekXg9tx7Pz7LFxezAyaH8bwcJHjyKC45v+2KwN+KFQ+5uXzPRZlPbNmC4a1JE+eLuu0A++CxzGe2TeB1hmil/6UE779R7/wO1UwpvLeKYdkHT2Q+4/fgO5KnednRdcPxlVtQnuPb15GipeC7LRO/lpQmXVNWhCvdc6RfLgtA51Q8PMZz724FIbg85zlH58pVOSwvdniXC2/tfGZ+TuXfDgqDLs8I/aImhBBCCCGEEH2GPtSEEEIIIYQQos/Qh5oQQgghhBBC9Bk91aix3qvVymKdv/mXfwbKHv7V3wWbl/wMY1cX/utzUDb0nGeCfckll2BFguVx/+3mW6Do2c/CpWZZCvCqX/jN9nYcYWz24CDG+s/NLRiS3cPPvfznoeRP3/kWsFstVCj84i+8OqgUBnZv2oTL4bYaTbDn3vzu9nYSsSaLdBH06T61NeujSoN1Yr1bIpt1Z6GOyNGS2LyUM+sbkiQ7tsLLazex3RsNtJvNbMn9hQXU57RI/9UizYkPtCCL2EWWcCQ8tW2pkvX5YJWW4ub7q+PJG4EWqNUgHUiC+zpaLjfUkeQ0aFHnMbnyD+GJi1NorDdc1zDdA8eY5+PmO2sB+FhegrzeQK1t3Mzas9FArdiRvfvA3rLnUrDnpjMd2oFp1ArVUzzX+Bguwb+5kvnMYozz0sOH7wd7A80nOzdlS1vXD6C+qXYCdTCO5pNG4F/1OmqUEkq5kdKxaYEeo4dTj8V0sSTQ2sVlfAY0WeviSEsVzDeJZ30bzkVRCW3XzNqrvoz93aBlzBfIPjk93d5eIh8dGUbh0LYJ1KgNWaa1rdK4rbMGI2WRSebvrP9lqZijhy00D6c9yOliSHsbtHuL7rfU47lngoRYC8EzZpHmgJRfyVJOp5LdV3QBplGop4+AXaujZvpslg2vBmOzMcW+heO6fAFq2OYCXesEaScTur3ZBO9/eTTT9Kbka9NHj4MdzeKzbMeVV7a3HyFtbY0ewHndVW91jJ0oqsfa69h5/26pltKgnHVmUcGy+Lx/XkbXTUfnO2ybeRrXxuM6PDdNNrzUv89p58LtYlFe0WL9rHfvLgY8NfpFTQghhBBCCCH6DH2oCSGEEEIIIUSfoQ81IYQQQgghhOgzzmsetQ0u0zfMbh+BssES6nCs1Tl3xfzffxRs1qhxTrJQwvTe930Yyt7/gY/gvqxtaWb2T/2/L8YqUlKUwUHMZ3XZ5Vl+kQfvR63HS172KiuiHGj0XvGql0BZibQMG0jPtz/J4uDzcc3Fepz7jmTtHm2mtiis8TkmFxccxExTu3vKr5HGFAgfOEArxTjnnEaL2ifUPmzahHlbFpuoN1ikGOpaoF9pUJ0bCeXvo6w4UaBJabhZKIs5NJvireMg70uFcxWldB3KEZME+YlaKdeZ9iWNEUrU0Hdi1pysMxwrjmZxvDr/PSsfV5+RemyThqEWYmgk06v4JdSQ7LsftWJNj/PH8OZsjpyokM6ojDqY8Q2bwN4+numQBndgHedmUeuRUi60+YOZNu7QQ3uhrEGax4FhnMfrrUyjt0S+1qR29sY5yDpr1M5KcLNGmjRHJMG1HT1CE4/94FmjFto0L0Wkd4srdK7lbI5YIn3s0MI82AstzOF4/GSmaUwMn60jEc49s1OHwF5Ks2sNVYZw3xbORdNTqJ1cWgrrydmJcgkuyQ7KSXfMOdhyGpswB1tKWsBeChzNbOsF+JyoHc1yo1WpKhMjeJ/DZbyvhbls7KYXXofndZ8Fu7l4DE9eD/piEPOKMTwFLuzPxu7hu3G+GLpsM9hj9Oq2ZSzz89FByn1WRv+59Cr068XlLYGF/Tg3g/vGdWy7XVsDje/JfVDma/zs6qy3Zn8pmv/PNYWSpjVWA+dTKsvlGqRLhTozmg99QnNczPN4cCxVOnakw+WbCtcioFeGlN4/fE7mH9SD33v4WNbWBro0bhu2mdA/zpXUUb+oCSGEEEIIIUSfoQ81IYQQQgghhOgz9KEmhBBCCCGEEH1GTzVqHNu7VM3iRmdnMWaa0n1ZqSAet0rxpRN1jF3eN4Kx/m/7vd9ub7/qF34ZyrzHAGtPeXH++I/f1N6OI4xz7Ra7/Es//7/a2ynF9b7ilahR43O97a2/396uN1DbksbYjXtf+qtgD8RB7ibSTCxSPPnQH2B+N19/Yrad0wX0kIJYX8/5lzi1D+f9KGd/n2hRPg3WaEWky0oDDVurifqjQYr7b1Euo3or0ArSeYcGUJ8yMoD6pKSR6UjqNdSUNKnOMfnDcDXTCPhFHBsp5dxLjfOJZI2ZUp1ZNZqknVWLnAcvl6BwnckpD9egM8jVNIz1pyLO2XRiEfOohbn0xshfmsuY3+zQQ/eCvcNnGtcLdm2BssENmDdtcBhzoY0OZD5RpjsaG0LfO7wX8zEd/WqmnVueRy2UK+MEMkex/8cWszEwR3q2JufGLNDLFuWyW2+apJcOp5uE9E8tyjVZp7yMqQt0aSUc4470X1EV+8XKgd5vAbU980eP4r4DOM+XLbsua3qXZ6bBPriIufG2TGZ6x53jqEdqzOH9TVF+tyiYixLKDZqTlVGfhnn08v1drBsJ8/n1VpGWZ4F0qjaY+czECOYcu2wCNZ4bRnCO+MrBzAeuuBLHfHnzNrDrC6g1tPlgftmAufIYv4h+PHso85EtF+J1h3fjPbBmes945tfVEq0XYOgTY9US2ZnvRcdPQFnzCNoDkxeC7Sezes7/+z1QVmqSRg2rTLo0Gkvl3r02F+Y3yycAxWLOW1hQ5rto80FjTs9uHtdFqdE4f2r3fKpZOa81kTvSF5yL3+NyOnS6avjew/q23LH0DhnULF+lM5uN9IuaEEIIIYQQQvQZ+lATQgghhBBCiD6jp6GPvNR9M/gV8J6lrVD2/LdiSOIjr3wz2N6C8K8BXOb4gVe8CezhP3kN2HGwnP1b3/pGLKOfVznci8MdoU4culEQCsarDXM9mCTJQruESwkAABhxSURBVB7iEoZynnzlTVjHMv682gy6uUppDo5RKGR6CO+htTkM8zp/y9TmfzLOfIn9iuuVFvzczEu+RtQeES3174NQpoU5XAJ7jJbXHhvCMJaBIJyRwxVLVezTCqVcaC5n9WpQ6EWLf3rnEIjlzHdauSVtKQSC2y5sH8/XocaLO4dp9NZX8vBYXFvkZecQktySwTTkF2sU7uWysOXSMIYQVWkp7tocpvBIHshC0mrzOF9u3IxLgI+NYxjdQhCPtLyA4ZhL0xjqduwghkxNz2YhU47CAOdpPjlBKQeml7P7ryXotxzqWLTscS70rWBuPdfEVQw/gxDfFoU+sp9FeM9R0A+OQh/TCOeA8gD2oS9n801zGUOY544eAXtscjfYl+7MliqPOSyyivcQ0fOjOpDd/wCFjNkkpSA5ieFoc0F4eNKk5yOFWxWFarFr5GcTXl49WBKc562ot3+fPjGHbbJ9IgsVHC2NQdkIpfQpV3G83XdftrT/Q5ux7PKnPRXs+D//Fuwjtx5ob28Zuhj3xWwe5gawjTZdHKT32ETpbtBtc4xUygWlGDYZGbaHP5kdO/OpA1g2hc/fDc99LtgH7818858/+l9Qxs9M59jBsnFbiilNyiCFJK8jawn55qL8sWEZHVuwLP7KDoHNaVUo3D+hQ8Pw59y7Gu56iuX5wyrRPMWh8/RQD5fgTz2lFOgSNhraLFPptjx/EWf6GqRf1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+o6caNQZiaCns894E44CnqxjcuaOWxWfz0qJNilVtvAz1X1f84W+0t0+SVoy1LBxT2wqWIs1podLOS5ubYXxxi5cqJjgWOdQ+LL8KNXfJMi3PXqDJSiJs14ve8Gqwv2w3gO2iYBloXxRrvr4U6eMc9RHrrKiLzQJ9WG7Z+BLFQZPGxHymuWjUcSnqxVmKmd80AfZouNRymfyK/J+1ZFEQ518p5ZwWzHApfzOz6WBJ/rTJminyUYr7TsKxRfvy2Cj+qw/1X48XzU7XoB/lcZvTtEAx33Wx7mo5SNkw5VArlg6hr43E6AMLs9kS2a15XMp/9sB+sIcGOJ4/267X8LzLNVw+vNnCtBIuWL5/oYllxxZRkzZVw/JaoIFMqG1yPcDCAdBUdNZbrDc10mWF+qeEtKY891iJ5szgptkFGyTuGBtBHeJwMBctnXwAj13GuWfu6MNguzgb91t37cQyhyKjmFLHlFvZdVOal44cnwJ732HUNzZ84NMsDOVlrnk+Cec5z/Myj7vOaVSYXmvUJidRq3zpeDaeRiPUIR6nsWkpzgkz99/R3v7Yw2+Dsiu+/xlgb7nu/wH76G1fzs7zCdSlbnwG6tvcBKZhGNwWvDec0+ZDX/NTmP5j7hOfa283H3wIyi6YuBzskkO929999N3t7Y8fxfFSL5MW0OO8FWqzc2mYWjhvryc53WaonSrSdJrldZ0wNxWnu8itKx9qxfgZR/Xg5frD5ex5DQgWeed0Z/CNUKxp5bkptDm1FKd04jmhKDUIt3O+JdeQUuE00S9qQgghhBBCCNFn6ENNCCGEEEIIIfoMfagJIYQQQgghRJ9xXvOouTDXCQV+fiXBPDDP+P3fAHvhpzM7reB5Y459J/v+n31de3tqI8Yq73jjK8EeSLHcfBb3OjKCsedLSxi7nCSd44uTBOtUo3xdpY98CuzGh2/OrhNRbjf63k4d5YyIMj3f6P/3K1D2ZbsKbGtgjHhsocaCYpN7+JmfT3MRxhBzjgzWQ3GMcaAVTDjOGW+qQgnvBgLNCfvZwtw02D5BrVirkeXNqg6jNqGV0zahnQS5SlhnltRQJ1SnPFmtQDcUF+QLMSNNmpklBWP0FMlIuthwcEHZ+hP6SF6yUqyHWkvNed8wB9d8DXNhNVKqyDDmSRoJ4vsb3Bl11JnVGqSrsHDuQd3UchOvWx5AHWsz0GUepzlumvLELbXIfwp8INeOBZoK3rd3WdTMYspxGOYNikijxq5Uoj4dibNzVQfx+eGGUdN6w/VPAvvqIG/W0Qf+E8q+9LnPgL3/MGoYj+27v729NH0SyiY27sB6GPb/9m3ZvHX0CGrf5k8cAzuex7moFLYI6X09S23L/A9ZO3t62LDtckmkQoEjFrU6p0JdF67fgUnK4iTTNk/N4jPj4EnUPc8cxPatt+5rb6e79kHZLftvBPsJHnOlbbzwK+3tY/swr1h0y3Gwx5/xHWC7oSy/mSuT7pI100zQ4L6J85SnvItT//1ptB/JNGrDw5jPsHrZC8D+wiz6/M1feW97+0Qdc1Kax7GXkjYqvKXIYZ1Zp9tLQr1Ut+dQce7SLvnZOK9caObOy+8UpBcN85mRjixfxYKcuLn3Da4j5YSFerHGletMxwbzeperFlKU924t6Bc1IYQQQgghhOgz9KEmhBBCCCGEEH2GPtSEEEIIIYQQos/ocR61zpotpkQB7F9o7gH7xnfc1N4++ao3QVlEugBKT2NxoHDYPEU6kZ/9HbDvncB47Gf90qva28cp3jqlHFwjFdSYDM1k+3/s994KZRcfQW1YahRIH4b6U9hrKaW8YaQx2fr217e3Pz2LeoRWiRsH2x36iHaNCmOgzzERa7jC/F64ay5HF+fbCGzOp8X5yxJHuTnizB+qFRw+juKcl0iDtBToRnxMfyMh/Ubi+X7DQtLFJKgTijg9UTnzw5Su20jQh+t0rlZQDc5NlMsZVaRJy+Ux6aXKKB9XHsao5/yY8yNSXSEdVMpzGl03Fzef2U3yvWad2wT7Yr4c6JtivFClhP5TIm1lKQ5yYdG+803UxUSUR215KculOE85G+strEeTbgHHxNrmC99h26y3GrVGfYH+JagNzx9klyJ8BlSD9iglOD+YQz30RRdfAPbTnnhNe/vkznEoGy9tBPuWT34c7AMnsrxZtZkTUHZiGp89kUMt0NRDmY4ooXxTTdIHD7NsJM7+oU7alQZrq1kvHCSXZD2bp7mWJTVxIDxNcvKb3upjx0n/dLyR+cTe46gXPHISfW2Z0l5WRrJn+71LqG/7qzv+EuyBoW8F+6qxS9rbO6/ANQCO770Z7Ol/+XO88HCWV21wN75DlLduwX05r+3hTP+2/PAR3HUetXHNFOeiPU94Qnvbj+H4qE1ibrQPfujfwb7nUJZbcqSC73GNOurHOc9m+AwNfdjMrDI8ZueLs9FHFb1v53RY9MyDvLWc05DO6xzrzAKd+xoF3/B+lquTkc31CA06Lz+XSWscPtbzVTz951iuyc9w7tEvakIIIYQQQgjRZ+hDTQghhBBCCCH6jJ6GPhYvF4rwT6RJC2MfPm+Pa2/f8KevhbKZn8PwxYFlDO9qFkVo0dq9e47jt+z9v/q77e3IOOwHj52mUA8LQhR3U7hig+IZ47TzT6Tcjp7Ca0b+4BfB/szSnvb2cplCpDpepft1e0mT0hnEQZhp7tfllH/WpvJg2zleWhiPpdXGrRmkZ6iRjw5TWGEpwXOHlqPQtIRCptIUfTYOlkSO6Y7ZDykqzlrBPyxQqGOthbE17MMQ3sjhp7nl1DsvZHuulqk9U/j6URCekfNrDqEoWMqYQyg4FQSbYchlvrXQf+Yb2DcLQVyhMwzdiejYmEKYBwezkKlqFeeLZcMlsmuU3qEZhNqSu+RSkOQSNgT/wEsi58NtTn/OO8W6zutGVMf2CZefdvy3Tg6F5SDNIHSaIoTMNzHsLaU+jrdl4Wdbh58CZdcs4bjefxyXI5+xr2bXXaT+X8Q6N1udfTjluTWipdqpX1pBWzWp/3mR8xadOw2exVGJpAC58Cqqcrg6Pzd0TwNnzQ4v4Z3uP5716/EFCjOmd5WEZBytONt/eXoKyu647cNgv+vIvWDfeOGPtre/9/nPgbKtT8el/R+4806w7WQQvnj4ISgaqdLcQ7KN+cXsfhc8prqIN+MbyCVPfTzYC8NXt7f3Th2Esls/hvKRuz6KKQcuqm5tbzfJt+b4OUgvheHjuFzG+xso4z2sKwWOnXsPLEhvslLe+djuhM9ylgIUhByamQtlBrwsfi7nT+dnhDOW5XAVuV6dJS55SULuZHb6dL7/tYSfFqFf1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+47xq1Aq/ErvEcjYti43/xOxlUHbj224C+5bXvAHsJx8Jlhu+eCeUJV20NElQ6aTzbmZWrLlodSxZgXV0rWB94snnPQPK7hvE5ZT3LkxiPSyLZR9YY4xsKJ1Lc+HS5093hNcmT2KNEWtBAh1WbrVYslmDlKZZzyWk5Yhpuf5SbsnorLxSQp1InJPKkYeE90Q6soScpUXCulorW/J4mTSYvJy6N47dDqrAsegUb17kWXmNUcHO6wBrx0Bnxjt38Wtc9Zc1Sad/n2mRHtBykkDsG/KPJvVNie6hEpS7mGPqaWliSv/ggrmH7zdiiVZ6+vqffNPkbrjjvr2ce1gDGrkg1QG1lW81aV/sJxcs9d3ivDGUFiGh9C/hPQ9swSXCxy7E5fknd+HS/pPLmf5tZhY1d9EsauFOnJgDO02z/vcJvTJwKgvyj2YwN7OmF+/uFFqyQFeUkEa5xEJccsSwlPUnuTG6zjzwMC5BfyLIyjA7j8vRu5RSvnhMtVPyWXmyiFrShTo+RD59AjVdB+7/h/b28v7PQ9lTvu+ZYF982beBPWmj7W3fxCX2/eIs2DwnjJSze2hU8V3FbcK+uPXB28H+3H+/u71916EDUHb0xD1gzx1AHxkaHGpvN4f4mUjrFvBzsECXWVvCdBbrSdE813UOzJUH+uiu6XE6H8tlPOZZd+YCHbPncZobxp3fKfLPANKdUXlYyu9x3TnzOQK0gPnCMzqnflETQgghhBBCiD5DH2pCCCGEEEII0WfoQ00IIYQQQggh+oyeatTidcp9wzfxucVLwd780leCvW13pg+6/+bPQNlTr7oS7E/d8nGwn/eyn2hvv/ft74Ky6669AuwTMxjr3xjIYshnvoy5SEoT41jHp2A+kalK9k39QLoJylyMeT3KDYwZXwtFeR5YRtVL4rjgbwoc50xah5jiopNAG5LLC8T5tgrkb55zn1HbNajOUVDeovjqmASAee1coKtLUffC9fCkumqEeiyKAY/oBnMavTSMt+b74XjzznlbcvHkPdY38tyTQt2K4vFPkRcm2O52H7k2CfqKfY//bJbX1nSuM9ciofZutDIdUokTWLnifoWYezpvzgdiznUT6iK69XlRHjXSAvbQfdzQCNjloFuSOuq7WFjYalJ5KdAKlrDtlhvYMbVFym939Fh7u7IR23mphNetXrAb7E1L2ZxRL6F2aaqO+bhOUp1rge8kDdRU8bhOuA+DOTAuYc61UhntKk+24Xl5XsrpktEO9Z+Ox0qPpdUJ6dCsntXVN1DjGBlql6sVerYH80eNjvUNPDaN8M3okcW97e2/Of4wlH1s//1gX3ot5lXbuTNbB2D75aiH3HXRhWCPONTV1R/JrvXIwTugbP/MI2DfcsfnwK61NrS3LyHt08gMjoEDbgPYteCZOjCEZb5Bz4MG6s5ckOUvabGWuJeTT+f3gnwesVwitYJjWQ/cRbNWeGzx8zN8/+JjWdOcpLx6Q+fcb900znGYK5U1rV3e+6BWxc16Bvq3taNf1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+47zmUTtXOhU+b4k0OvXJ68D+79pge7t6Je47+PSrwP6VX3w52Pfdd3d7+/X//n4oa82jNuwjN98M9vNf8D1ZHRt43X2HUSfQGt0C9l/89Qfb2z4m3UMXQh3S+cx9djawRs0XBBFHrtgOc+Xl83uRXof+lBGHGhRqyiTl62B5mG+EY7E9J+XjuGjI2sf6JIrdjnFY+yAZHueE4pwnOc1eMJZYz8Z2TtsV1DM11qidfr6tc0GJtENpcN8tzv3FLlGg2+SSmPPA5LQzabAvaSdzurOCsdolXRvnYLPgupyfi+P1HdUrzBXGvZbLg1PQVmeTyob9JZdzax2plCi3VdAePAZieqQ643pn25xHr0b6ry/ehXmiyoEk5/FP2gVlvoy6oPGd14N9+eBEe3uphXqce/ehXmkxQa3cUqDDS1mv0yWnUhwom6sxP6eL8zKmoOfk+YPfJTofy3nTWNO73iyRjjEuZbnEXETKb36mJKhTTANdWovnarot1qk2g90b1Bezhw+B/ZUjHwTbKpn+bXInajZvvPHJYF+14xqw5++8rb19z74vQNkjLfT5yjDmB7x0a+a3Ywu47/QcahyHJlC7P78UaHoTzN+2dfMOsHduw/UESpYlu1tcOAFlLumd//D7WtGzqEiTxsfyuO3+bh6W83tA5+d+zqaxx/pRHgBw5i4PkFxpcE9FzyUzM5/TZWf1zD2Hc+9mXWty1ugXNSGEEEIIIYToM/ShJoQQQgghhBB9xnkNfSz8GfesKF5I3vnsJ/SIlgz+x0/cDfY/feJesB/8+Ifa29MNDCEaruB3b3UQw1H+8N3vaW9vvXAPlO256gawc1/QwS3xMundWEu44/r1ydmR852gCRwvc05xhBxi1GxmYYcuwv6PaHlxiqihpVgp3IZDEnPtHvycnvv5nJe6x3qEP93zUuXd1o8Nl1TnqB8O1+T7jYLQHA5xyC/jTtVy4WbnpXN7QcQhJEE/c1qFXJsUNC8v+x9TKBOHxoXX4tAkDsnKr4JcELJLcPgJxEWxE3BYGaedCI/NhSDngiE72rmUAtywEY+BzveYS22wjuzaiGFSzdZCe3thEcMIl2oY5ubLQ2CXgntqUIiho+fWXAvtvWn2PKksYLuPDqBdGsR5bWrqZHv7vq8+CGUn5hbATrhPw1QWpwj0Cenc+6fYN1eYy0mSbdIcnvPZVud0FRxSnPP/daZewfcAC8IOS1Xs44SW3K9TOGwp8JG4hOcNU3CYmbU4UDkYXvxusmEMQw5byxgKeLI+096erWMagENHpsG+7lL0+XRDtpz/vbOU+mEDhiQ+lZ57uxey6+5NcX44QWNrOddW2bU2lPFd7cLtmA4pJrmAK2XnLvlRKGvMLVmvWMv7WP6dAsE5s3jnXGg57M4PplxN8LrwnMu9YJDZOaVR1+cHPU8h/DkXQkrPsaJwzrQ43HstkY5n+n6tX9SEEEIIIYQQos/Qh5oQQgghhBBC9Bn6UBNCCCGEEEKIPqOnGjXmfOmhwiXH0xhjtaMSxl9zHOyTnvn89narhEvnbqxiLOtnb0V92zc+9enZdR1e10cYQ51fBrugrTx9b7vOMfjdlnvNhxv3x7c831G43HquZXJLzHdeBjqh+OOEl5fmWO3wWFqqOklYG9dZS5ZbTj3CoViOud2Dpe5ZQ0R75jMOBOkZchfmVAZcfsrNlfN2iRkHNZ9j/VFv/YrbyAd9w3rAOOL74qM7aw1blJPBUV+FIfgc684awNx628H+XccxHRv6eUJjOirSsxmmHOCUAilrONk308515iXVqdlBDsdDieuxnmzejcuNHzh8R3t7oYbPgE07tuKxEzvBHqlm2pfDR1ErVjtwFOxKhPqdDaPb29utZBDKZhZQK3f4gTvB/srtn21v7zt4EsrmadnznEYNZivWmbMPs28FaSFIk5cmNCeUeI4Ixxlep9FErVOa07oEZeQ75R76jpnZhTvQBw4dytIhDBimadm4EbVTddI8zi5kbVin50/K4yuXOiMrbzXQbxcW8djhITz3pbu3tbfLwxdD2fgQaji30TL59eAda9hvhLK0jr52ZAm1cQsnM/sEzdMjGzCF0QU7sZ2P7f9qVufmESibP4LjpUUuMe2zdt46ifq92jweu54UzvNc1uVcOIqLlt+3U8jew/ct0lLzhbheUairLHUsW7E765T5OrnnCetWk2xsJQmlQ8q9m2F5qKnPa8WtEGjnc5SSrD/ewoUQQgghhBBCtNGHmhBCCCGEEEL0GfpQE0IIIYQQQog+w51pzKQQQgghhBBCiPVBv6gJIYQQQgghRJ+hDzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+Qx9qQgghhBBCCNFn6ENNCCGEEEIIIfoMfagJIYQQQgghRJ+hDzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+Qx9qQgghhBBCCNFn6ENNCCGEEEIIIfoMfagJIYQQQgghRJ+hDzUhhBBCCCGE6DP0oSaEEEIIIYQQfYY+1IQQQgghhBCiz9CHmhBCCCGEEEL0GfpQE0IIIYQQQog+Qx9qQgghhBBCCNFn/F/BJZA9axvaqAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0be67c9550>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0bf95ef898>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0bf9cfe2b0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0bf9b162b0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0be41b69b0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "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",
    "k_size = 5\n",
    "softmax_logits = tf.nn.softmax(logits)\n",
    "top_k = tf.nn.top_k(softmax_logits, k=k_size)\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    saver.restore(sess, model_file)\n",
    "    my_softmax_logits = sess.run(softmax_logits, feed_dict={traffic_sign_x: features_train})\n",
    "    my_top_k = sess.run(top_k, feed_dict={traffic_sign_x: features_train})\n",
    "\n",
    "    for i in range(5):\n",
    "        softmax_image = {}\n",
    "        title = {}\n",
    "        \n",
    "        softmax_image[0] = my_images[i]\n",
    "        title[0] = \"My Image\"\n",
    "        \n",
    "        for j in range(k_size):\n",
    "            softmax_image[j+1] = X_valid[np.argwhere(y_valid == my_top_k[1][i][j])[0]].squeeze()\n",
    "            title[j+1] = 'Probability : ({:.0f})'.format(my_top_k[0][i][j])\n",
    "        \n",
    "        fig,axes = plt.subplots(1, 6, figsize=(15,15))\n",
    "        for row in range(1):\n",
    "            for col in range(6):\n",
    "                a = axes[col]\n",
    "                a.axis('off')\n",
    "                a.imshow(softmax_image[(row*5) + col])\n",
    "                a.set_title(title[(row*5) + col], fontsize=12)\n",
    "        plt.savefig('examples/test_image_'+str(i))\n",
    "        plt.show()"
   ]
  },
  {
   "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": 14,
   "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\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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
}