Traffic_Sign_Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Self-Driving Car Engineer Nanodegree\n",
    "\n",
    "## Deep Learning\n",
    "\n",
    "## Project: Build a Traffic Sign Recognition Classifier\n",
    "\n",
    "In this notebook, a template is provided for you to implement your functionality in stages, which is required to successfully complete this project. If additional code is required that cannot be included in the notebook, be sure that the Python code is successfully imported and included in your submission if necessary. \n",
    "\n",
    "> **Note**: Once you have completed all of the code implementations, you need to finalize your work by exporting the iPython Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to  \\n\",\n",
    "    \"**File -> Download as -> HTML (.html)**. Include the finished document along with this notebook as your submission. \n",
    "\n",
    "In addition to implementing code, there is a writeup to complete. The writeup should be completed in a separate file, which can be either a markdown file or a pdf document. There is a [write up template](https://github.com/udacity/CarND-Traffic-Sign-Classifier-Project/blob/master/writeup_template.md) that can be used to guide the writing process. Completing the code template and writeup template will cover all of the [rubric points](https://review.udacity.com/#!/rubrics/481/view) for this project.\n",
    "\n",
    "The [rubric](https://review.udacity.com/#!/rubrics/481/view) contains \"Stand Out Suggestions\" for enhancing the project beyond the minimum requirements. The stand out suggestions are optional. If you decide to pursue the \"stand out suggestions\", you can include the code in this Ipython notebook and also discuss the results in the writeup file.\n",
    "\n",
    "\n",
    ">**Note:** Code and Markdown cells can be executed using the **Shift + Enter** keyboard shortcut. In addition, Markdown cells can be edited by typically double-clicking the cell to enter edit mode."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "---\n",
    "## Step 0: Load The Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import cv2\n",
    "import math\n",
    "import numpy as np\n",
    "from sklearn.utils import shuffle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Load pickled data\n",
    "import pickle\n",
    "\n",
    "# TODO: Fill this in based on where you saved the training and testing data\n",
    "\n",
    "training_file = 'train.p'\n",
    "validation_file='valid.p'\n",
    "testing_file = 'test.p'\n",
    "\n",
    "with open(training_file, mode='rb') as f:\n",
    "    train = pickle.load(f)\n",
    "with open(validation_file, mode='rb') as f:\n",
    "    valid = pickle.load(f)\n",
    "with open(testing_file, mode='rb') as f:\n",
    "    test = pickle.load(f)\n",
    "    \n",
    "X_train, y_train = train['features'], train['labels']\n",
    "X_valid, y_valid = valid['features'], valid['labels']\n",
    "X_test, y_test = test['features'], test['labels']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Provide a Basic Summary of the Data Set Using Python, Numpy and/or Pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "43"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_label={}\n",
    "for v in y_train:\n",
    "    dict_label[v]=0\n",
    "for v in y_test:\n",
    "    dict_label[v]=0\n",
    "for v in y_valid:\n",
    "    dict_label[v]=0\n",
    "len(dict_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training examples = 34799\n",
      "Number of testing examples = 12630\n",
      "Image data shape = (32, 32, 1)\n",
      "Number of classes = 43\n"
     ]
    }
   ],
   "source": [
    "### Replace each question mark with the appropriate value. \n",
    "### Use python, pandas or numpy methods rather than hard coding the results\n",
    "\n",
    "# TODO: Number of training examples\n",
    "n_train = X_train.shape[0]\n",
    "\n",
    "# TODO: Number of testing examples.\n",
    "n_test = X_test.shape[0]\n",
    "\n",
    "# TODO: What's the shape of an traffic sign image?\n",
    "image_shape = (X_train.shape[1],X_train.shape[2],1)\n",
    "\n",
    "# TODO: How many unique classes/labels there are in the dataset.\n",
    "dict_label={}\n",
    "for v in y_train:\n",
    "    dict_label[v]=0\n",
    "for v in y_test:\n",
    "    dict_label[v]=0\n",
    "for v in y_valid:\n",
    "    dict_label[v]=0\n",
    "\n",
    "n_classes = len(dict_label)\n",
    "\n",
    "print(\"Number of training examples =\", n_train)\n",
    "print(\"Number of testing examples =\", n_test)\n",
    "print(\"Image data shape =\", image_shape)\n",
    "print(\"Number of classes =\", n_classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Include an exploratory visualization of the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "### Data exploration visualization code goes here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import matplotlib.pyplot as plt\n",
    "# Visualizations will be shown in the notebook.\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Container object of 43 artists>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAD8CAYAAACRkhiPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYYaAJDXMEJCkhv1/jbAomGDjeJgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6df26d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_label=np.zeros(len(dict_label))\n",
    "for v in y_train:\n",
    "    train_label[v]+=1\n",
    "valid_label=np.zeros(len(dict_label))\n",
    "for v in y_valid:\n",
    "    valid_label[v]+=1\n",
    "plt.bar(range(len(train_label)),train_label,alpha=0.3,color='b')\n",
    "plt.bar(range(len(valid_label)),valid_label,alpha=0.3,color='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# try to generate fake data\n",
    "# idea: affine transform to image\n",
    "def generateFake(img):\n",
    "    q0 = math.ceil(image_shape[0]/4)\n",
    "    q1 = math.ceil(image_shape[1]/4)\n",
    "    r=np.random.normal(0, 0.5,6)\n",
    "    pts1 = np.float32([[q0     ,q1]     ,[q0     ,q1*3]     ,[q0*3     ,q1*3]])\n",
    "    pts2 = np.float32([[q0+r[0],q1+r[1]],[q0+r[2],q1*3+r[3]],[q0*3+r[4],q1*3+r[5]]])\n",
    "    \n",
    "    dst = np.zeros((image_shape[0],image_shape[1],3), np.uint8)\n",
    "    mean =cv2.mean(img)\n",
    "    dst[:,:] =(mean[0],mean[1],mean[2])\n",
    "    Mat = cv2.getAffineTransform(pts1,pts2)\n",
    "\n",
    "    return cv2.warpAffine(img,Mat,(image_shape[0],image_shape[1]),dst, borderMode=cv2.BORDER_TRANSPARENT)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(47.5390625, 42.814453125, 41.2919921875, 0.0)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv2.mean(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6da898d0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFMr/A2oQYr/XbnhvAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6db6de80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6d9fbdd8>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6da23e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(generateFake(X_train[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "train_label=np.zeros(len(dict_label))\n",
    "X_train_fake=[]\n",
    "y_train_fake=[]\n",
    "for v in y_train:\n",
    "    train_label[v]+=1\n",
    "for c in range(n_classes):\n",
    "    # amplify datas\n",
    "    data_num = 0\n",
    "    indexes = [i for i, lbl in enumerate(y_train) if lbl == c]\n",
    "    while data_num < 3000:\n",
    "        index=np.random.choice(indexes)\n",
    "        X_train_fake.append(generateFake(X_train[index]))\n",
    "        y_train_fake.append(c)\n",
    "        data_num+=1\n",
    "        \n",
    "valid_label=np.zeros(len(dict_label))\n",
    "X_valid_fake=[]\n",
    "y_valid_fake=[]\n",
    "for v in y_valid:\n",
    "    valid_label[v]+=1\n",
    "for c in range(n_classes):\n",
    "    # amplify datas\n",
    "    data_num = 0\n",
    "    indexes = [i for i, lbl in enumerate(y_valid) if lbl == c]\n",
    "    while data_num < 500:\n",
    "        index=np.random.choice(indexes)\n",
    "        X_valid_fake.append(generateFake(X_valid[index]))\n",
    "        y_valid_fake.append(c)\n",
    "        data_num+=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6cdae518>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6d9984a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[2295])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6cd149e8>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwS9Eoij4hUiU/wdfn2tkLAwctQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6cdd00b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train_fake[2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Container object of 43 artists>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAD8CAYAAAB+UHOxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJKlTBoAkdcoAkKRO/R/jkkZ7TxUEhgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6da590f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_label=np.zeros(len(dict_label))\n",
    "for v in y_train_fake:\n",
    "    train_label[v]+=1\n",
    "valid_label=np.zeros(len(dict_label))\n",
    "for v in y_valid_fake:\n",
    "    valid_label[v]+=1\n",
    "plt.bar(range(len(train_label)),train_label,alpha=0.3,color='b')\n",
    "plt.bar(range(len(valid_label)),valid_label,alpha=0.3,color='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Pre-process the Data Set (normalization, grayscale, etc.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Use the code cell (or multiple code cells, if necessary) to implement the first step of your project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6a653ac8>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6cd29978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Preprocess the data here. Preprocessing steps could include normalization, converting to grayscale, etc.\n",
    "### Feel free to use as many code cells as needed.\n",
    "plt.imshow(X_train_fake[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+UUKpeQXKZSSX6RQSn6RQv0fEDgca5OtH1kAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6a5e69e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# convert to grayscale\n",
    "gray=np.zeros(image_shape)\n",
    "gray=cv2.cvtColor(X_train_fake[0],cv2.COLOR_RGB2GRAY)\n",
    "plt.imshow(gray)\n",
    "plt.gray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb6a524ac8>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6a5deda0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# normalize image\n",
    "norm=np.zeros(image_shape)\n",
    "norm = cv2.normalize(gray,norm,alpha=0, beta=255, norm_type=cv2.NORM_MINMAX)\n",
    "#cv2.normalize(gray, norm, 0, 255, cv2.NORM_MINMAX)\n",
    "plt.imshow(norm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def preProcessImage(src):\n",
    "    gray=np.zeros(image_shape)\n",
    "    gray=cv2.cvtColor(src,cv2.COLOR_RGB2GRAY)\n",
    "    norm=np.zeros(image_shape)\n",
    "    norm = cv2.normalize(gray,norm,alpha=0, beta=1.0, norm_type=cv2.NORM_MINMAX,dtype=cv2.CV_32F)\n",
    "    return norm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_train_prepro=[]\n",
    "for im in X_train:\n",
    "    X_train_prepro.append(np.reshape(preProcessImage(im),(32,32,1)))\n",
    "#plt.imshow(X_train_prepro[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_test_prepro=[]\n",
    "for im in X_test:\n",
    "    X_test_prepro.append(np.reshape(preProcessImage(im),(32,32,1)))\n",
    "#plt.imshow(X_test_prepro[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_valid_prepro=[]\n",
    "for im in X_valid:\n",
    "    X_valid_prepro.append(np.reshape(preProcessImage(im),(32,32,1)))\n",
    "#plt.imshow(X_valid_prepro[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_train_fake_prepro=[]\n",
    "for im in X_train_fake:\n",
    "    X_train_fake_prepro.append(np.reshape(preProcessImage(im),(32,32,1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_valid_fake_prepro=[]\n",
    "for im in X_valid_fake:\n",
    "    X_valid_fake_prepro.append(np.reshape(preProcessImage(im),(32,32,1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Model Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "### Define your architecture here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "import tensorflow as tf\n",
    "from tensorflow.contrib.layers import flatten\n",
    "\n",
    "EPOCHS = 10\n",
    "BATCH_SIZE = 100\n",
    "def LeNet(x):    \n",
    "    # Hyperparameters\n",
    "    mu = 0\n",
    "    sigma = 0.1\n",
    "\n",
    "    # Layer 1: Convolutional. Input = 32x32x1. Output = 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",
    "\n",
    "    # Activation.\n",
    "    conv1 = tf.nn.relu(conv1)\n",
    "\n",
    "    # Pooling. Input = 28x28x6. Output = 14x14x6.\n",
    "    conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "    #conv1 = tf.nn.dropout(conv1,0.7)\n",
    "\n",
    "    # Layer 2: Convolutional. Output = 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",
    "\n",
    "    # Activation.\n",
    "    conv2 = tf.nn.relu(conv2)\n",
    "\n",
    "    # Pooling. Input = 10x10x16. Output = 5x5x16.\n",
    "    conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "    #conv2 = tf.nn.dropout(conv2,0.5)\n",
    "\n",
    "    # Flatten. Input = 5x5x16. Output = 400.\n",
    "    fc0   = flatten(conv2)\n",
    "\n",
    "    # Layer 3: Fully Connected. Input = 400. Output = 120.\n",
    "    fc1_W = tf.Variable(tf.truncated_normal(shape=(400, 120), mean = mu, stddev = sigma))\n",
    "    fc1_b = tf.Variable(tf.zeros(120))\n",
    "    fc1   = tf.matmul(fc0, fc1_W) + fc1_b\n",
    "\n",
    "    # Activation.\n",
    "    fc1    = tf.nn.relu(fc1)\n",
    "\n",
    "    # Layer 4: Fully Connected. Input = 120. Output = 84.\n",
    "    fc2_W  = tf.Variable(tf.truncated_normal(shape=(120, 84), mean = mu, stddev = sigma))\n",
    "    fc2_b  = tf.Variable(tf.zeros(84))\n",
    "    fc2    = tf.matmul(fc1, fc2_W) + fc2_b\n",
    "\n",
    "    # Activation.\n",
    "    fc2    = tf.nn.relu(fc2)\n",
    "\n",
    "    # Layer 5: Fully Connected. Input = 84. Output = n_classes.\n",
    "    fc3_W  = tf.Variable(tf.truncated_normal(shape=(84, n_classes), mean = mu, stddev = sigma))\n",
    "    fc3_b  = tf.Variable(tf.zeros(n_classes))\n",
    "    logits = tf.matmul(fc2, fc3_W) + fc3_b\n",
    "\n",
    "    return logits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def DeepNet(x):    \n",
    "    # Hyperparameters\n",
    "    mu = 0\n",
    "    sigma = 0.1\n",
    "\n",
    "    # Layer 1: Convolutional. Input = 32x32x1. Output = 30x30x6.\n",
    "    conv1_W = tf.Variable(tf.truncated_normal(shape=(3, 3, 1, 6), mean = mu, stddev = sigma))\n",
    "    conv1_b = tf.Variable(tf.zeros(6))\n",
    "    global conv1\n",
    "    conv1 = tf.nn.conv2d(x, conv1_W, strides=[1, 1, 1, 1], padding='VALID') + conv1_b\n",
    "\n",
    "    # Activation.\n",
    "    conv1 = tf.nn.relu(conv1)\n",
    "\n",
    "    # Pooling. Input = 30x30x6. Output = 15x15x6.\n",
    "    #conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "\n",
    "    # Layer 2: Convolutional. Output = 28x28x16.\n",
    "    conv2_W = tf.Variable(tf.truncated_normal(shape=(3, 3, 6, 16), mean = mu, stddev = sigma))\n",
    "    conv2_b = tf.Variable(tf.zeros(16))\n",
    "    global conv2\n",
    "    conv2 = tf.nn.conv2d(conv1, conv2_W, strides=[1, 1, 1, 1], padding='VALID') + conv2_b\n",
    "\n",
    "    # Activation.\n",
    "    conv2 = tf.nn.relu(conv2)\n",
    "\n",
    "    # Pooling. Input = 28x28x16. Output = 14x14x16.\n",
    "    conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "\n",
    "    # Layer 3: Convolutional. Output = 12x12x26.\n",
    "    conv3_W = tf.Variable(tf.truncated_normal(shape=(3, 3, 16, 26), mean = mu, stddev = sigma))\n",
    "    conv3_b = tf.Variable(tf.zeros(26))\n",
    "    global conv3\n",
    "    conv3 = tf.nn.conv2d(conv2, conv3_W, strides=[1, 1, 1, 1], padding='VALID') + conv3_b\n",
    "\n",
    "    # Activation.\n",
    "    conv3 = tf.nn.relu(conv3)\n",
    "\n",
    "    # Layer 4: Convolutional. Output = 10x10x36.\n",
    "    conv4_W = tf.Variable(tf.truncated_normal(shape=(3, 3, 26, 36), mean = mu, stddev = sigma))\n",
    "    conv4_b = tf.Variable(tf.zeros(36))\n",
    "    global conv4\n",
    "    conv4 = tf.nn.conv2d(conv3, conv4_W, strides=[1, 1, 1, 1], padding='VALID') + conv4_b\n",
    "\n",
    "    # Activation.\n",
    "    conv4 = tf.nn.relu(conv4)\n",
    "    # Pooling. Input = 11x11x36. Output = 5x5x36.\n",
    "    conv4 = tf.nn.max_pool(conv4, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
    "\n",
    "    # Layer 5: Convolutional. Output = 1x1x4000.\n",
    "    conv5_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 36, 4000), mean = mu, stddev = sigma))\n",
    "    conv5_b = tf.Variable(tf.zeros(4000))\n",
    "    global conv5\n",
    "    conv5 = tf.nn.conv2d(conv4, conv5_W, strides=[1, 1, 1, 1], padding='VALID') + conv5_b\n",
    "\n",
    "    # Activation.\n",
    "    conv5 = tf.nn.relu(conv5)\n",
    "    \n",
    "    # Flatten. Input = 1x1x400. Output = 4000.\n",
    "    fc0   = flatten(conv5)\n",
    "\n",
    "    # Layer 3: Fully Connected. Input = 4000. Output = 2000.\n",
    "    fc1_W = tf.Variable(tf.truncated_normal(shape=(4000, 2000), mean = mu, stddev = sigma))\n",
    "    fc1_b = tf.Variable(tf.zeros(2000))\n",
    "    fc1   = tf.matmul(fc0, fc1_W) + fc1_b\n",
    "\n",
    "    # Activation.\n",
    "    fc1    = tf.nn.relu(fc1)\n",
    "    fc1 = tf.nn.dropout(fc1, keep_prob1)\n",
    "    \n",
    "    # Layer 4: Fully Connected. Input = 2000. Output = 1000\n",
    "    fc2_W  = tf.Variable(tf.truncated_normal(shape=(2000, 1000), mean = mu, stddev = sigma))\n",
    "    fc2_b  = tf.Variable(tf.zeros(1000))\n",
    "    fc2    = tf.matmul(fc1, fc2_W) + fc2_b\n",
    "\n",
    "    # Activation.\n",
    "    fc2    = tf.nn.relu(fc2)\n",
    "    fc2 = tf.nn.dropout(fc2, keep_prob2)\n",
    "\n",
    "    # Layer 4: Fully Connected. Input = 1000. Output = 500\n",
    "    fc3_W  = tf.Variable(tf.truncated_normal(shape=(1000, 500), mean = mu, stddev = sigma))\n",
    "    fc3_b  = tf.Variable(tf.zeros(500))\n",
    "    fc3    = tf.matmul(fc2, fc3_W) + fc3_b\n",
    "\n",
    "    # Activation.\n",
    "    fc3    = tf.nn.relu(fc3)\n",
    "    fc3 = tf.nn.dropout(fc3, keep_prob3)\n",
    "\n",
    "    # Layer 4: Fully Connected. Input = 500. Output = 100\n",
    "    fc4_W  = tf.Variable(tf.truncated_normal(shape=(500, 100), mean = mu, stddev = sigma))\n",
    "    fc4_b  = tf.Variable(tf.zeros(100))\n",
    "    fc4    = tf.matmul(fc3, fc4_W) + fc4_b\n",
    "\n",
    "    # Activation.\n",
    "    fc4    = tf.nn.relu(fc4)\n",
    "    fc4 = tf.nn.dropout(fc4, keep_prob4)\n",
    "\n",
    "    # Layer 5: Fully Connected. Input = 84. Output = n_classes.\n",
    "    fc5_W  = tf.Variable(tf.truncated_normal(shape=(100, n_classes), mean = mu, stddev = sigma))\n",
    "    fc5_b  = tf.Variable(tf.zeros(n_classes))\n",
    "    logits = tf.matmul(fc4, fc5_W) + fc5_b\n",
    "\n",
    "    return logits"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Train, Validate and Test the Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "### Train your model here.\n",
    "### Calculate and report the accuracy on the training and validation set.\n",
    "### Once a final model architecture is selected, \n",
    "### the accuracy on the test set should be calculated and reported as well.\n",
    "### Feel free to use as many code cells as needed.\n",
    "x = tf.placeholder(tf.float32, (None, 32, 32,1))\n",
    "y = tf.placeholder(tf.int32, (None))\n",
    "keep_prob1 = tf.placeholder(tf.float32)\n",
    "keep_prob2 = tf.placeholder(tf.float32)\n",
    "keep_prob3 = tf.placeholder(tf.float32)\n",
    "keep_prob4 = tf.placeholder(tf.float32)\n",
    "one_hot_y = tf.one_hot(y, n_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "rate = 0.001\n",
    "\n",
    "logits = DeepNet(x)\n",
    "cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits, one_hot_y)\n",
    "loss_operation = tf.reduce_mean(cross_entropy)\n",
    "\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate = rate)\n",
    "training_operation = optimizer.minimize(loss_operation)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Model Evaluation\n",
    "correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(one_hot_y, 1))\n",
    "accuracy_operation = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))\n",
    "saver = tf.train.Saver()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def evaluate(X_data, y_data):\n",
    "    num_examples = len(X_data)\n",
    "    total_accuracy = 0\n",
    "    sess = tf.get_default_session()\n",
    "    for offset in range(0, num_examples, BATCH_SIZE):\n",
    "        batch_x, batch_y = X_data[offset:offset+BATCH_SIZE], y_data[offset:offset+BATCH_SIZE]\n",
    "        accuracy = sess.run(accuracy_operation, feed_dict={x: batch_x, y: batch_y, keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "        total_accuracy += (accuracy * len(batch_x))\n",
    "    return total_accuracy / num_examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "\n",
      "EPOCH 1 ...\n",
      "Validation Accuracy = 0.954\n",
      "\n",
      "EPOCH 2 ...\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 3 ...\n",
      "Validation Accuracy = 0.945\n",
      "\n",
      "EPOCH 4 ...\n",
      "Validation Accuracy = 0.941\n",
      "\n",
      "EPOCH 5 ...\n",
      "Validation Accuracy = 0.959\n",
      "\n",
      "EPOCH 6 ...\n",
      "Validation Accuracy = 0.957\n",
      "\n",
      "EPOCH 7 ...\n",
      "Validation Accuracy = 0.962\n",
      "\n",
      "EPOCH 8 ...\n",
      "Validation Accuracy = 0.965\n",
      "\n",
      "EPOCH 9 ...\n",
      "Validation Accuracy = 0.962\n",
      "\n",
      "EPOCH 10 ...\n",
      "Validation Accuracy = 0.955\n",
      "\n",
      "Model saved\n",
      "Test Accuracy = 0.948\n"
     ]
    }
   ],
   "source": [
    "# Training the Modle\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    num_examples = len(X_train_fake_prepro)\n",
    "    \n",
    "    print(\"Training...\")\n",
    "    print()\n",
    "    for i in range(EPOCHS):\n",
    "        X_train_fake_prepro, y_train_fake = shuffle(X_train_fake_prepro, y_train_fake)\n",
    "        for offset in range(0, num_examples, BATCH_SIZE):\n",
    "            end = offset + BATCH_SIZE\n",
    "            batch_x, batch_y = X_train_fake_prepro[offset:end], y_train_fake[offset:end]\n",
    "            sess.run(training_operation,feed_dict={x: batch_x, y: batch_y, keep_prob1: 0.8,keep_prob2: 0.8,keep_prob3:0.8,keep_prob4:0.5})\n",
    "            \n",
    "        validation_accuracy = evaluate(X_valid_fake_prepro, y_valid_fake)\n",
    "        print(\"EPOCH {} ...\".format(i+1))\n",
    "        print(\"Validation Accuracy = {:.3f}\".format(validation_accuracy))\n",
    "        print()\n",
    "        \n",
    "    saver.save(sess, './Deepnet')\n",
    "    print(\"Model saved\")\n",
    "    test_accuracy = evaluate(X_test_prepro, y_test)\n",
    "    print(\"Test Accuracy = {:.3f}\".format(test_accuracy))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Load and Output the Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAABcCAYAAAB+6068AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5B0FBuzqYwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb377a7dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import glob\n",
    "### Load the images and plot them here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "img_flist=glob.glob('img/*')\n",
    "fig, ax = plt.subplots(1,len(img_flist))\n",
    "img=np.zeros([32,32])\n",
    "img_list=np.zeros([len(img_flist),32,32,1])\n",
    "for im in range(len(img_flist)):\n",
    "    ax[im].axis('off')\n",
    "    img = preProcessImage(cv2.imread(img_flist[im]))\n",
    "    ax[im].imshow(img)\n",
    "    img_list[im] = np.reshape(img,(32,32,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Predict the Sign Type for Each Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ClassId</th>\n",
       "      <th>SignName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Speed limit (20km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Speed limit (30km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Speed limit (50km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Speed limit (60km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>Speed limit (70km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>Speed limit (80km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>End of speed limit (80km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>Speed limit (100km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>Speed limit (120km/h)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>No passing</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>No passing for vehicles over 3.5 metric tons</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>Right-of-way at the next intersection</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>12</td>\n",
       "      <td>Priority road</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>13</td>\n",
       "      <td>Yield</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>14</td>\n",
       "      <td>Stop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>15</td>\n",
       "      <td>No vehicles</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>16</td>\n",
       "      <td>Vehicles over 3.5 metric tons prohibited</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>17</td>\n",
       "      <td>No entry</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>18</td>\n",
       "      <td>General caution</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>19</td>\n",
       "      <td>Dangerous curve to the left</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>20</td>\n",
       "      <td>Dangerous curve to the right</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>21</td>\n",
       "      <td>Double curve</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>22</td>\n",
       "      <td>Bumpy road</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>23</td>\n",
       "      <td>Slippery road</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>24</td>\n",
       "      <td>Road narrows on the right</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25</td>\n",
       "      <td>Road work</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>26</td>\n",
       "      <td>Traffic signals</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>27</td>\n",
       "      <td>Pedestrians</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>28</td>\n",
       "      <td>Children crossing</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>29</td>\n",
       "      <td>Bicycles crossing</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>30</td>\n",
       "      <td>Beware of ice/snow</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>31</td>\n",
       "      <td>Wild animals crossing</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>32</td>\n",
       "      <td>End of all speed and passing limits</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>33</td>\n",
       "      <td>Turn right ahead</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>34</td>\n",
       "      <td>Turn left ahead</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>35</td>\n",
       "      <td>Ahead only</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>36</td>\n",
       "      <td>Go straight or right</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>37</td>\n",
       "      <td>Go straight or left</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>38</td>\n",
       "      <td>Keep right</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>39</td>\n",
       "      <td>Keep left</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>40</td>\n",
       "      <td>Roundabout mandatory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>41</td>\n",
       "      <td>End of no passing</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>42</td>\n",
       "      <td>End of no passing by vehicles over 3.5 metric ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    ClassId                                           SignName\n",
       "0         0                               Speed limit (20km/h)\n",
       "1         1                               Speed limit (30km/h)\n",
       "2         2                               Speed limit (50km/h)\n",
       "3         3                               Speed limit (60km/h)\n",
       "4         4                               Speed limit (70km/h)\n",
       "5         5                               Speed limit (80km/h)\n",
       "6         6                        End of speed limit (80km/h)\n",
       "7         7                              Speed limit (100km/h)\n",
       "8         8                              Speed limit (120km/h)\n",
       "9         9                                         No passing\n",
       "10       10       No passing for vehicles over 3.5 metric tons\n",
       "11       11              Right-of-way at the next intersection\n",
       "12       12                                      Priority road\n",
       "13       13                                              Yield\n",
       "14       14                                               Stop\n",
       "15       15                                        No vehicles\n",
       "16       16           Vehicles over 3.5 metric tons prohibited\n",
       "17       17                                           No entry\n",
       "18       18                                    General caution\n",
       "19       19                        Dangerous curve to the left\n",
       "20       20                       Dangerous curve to the right\n",
       "21       21                                       Double curve\n",
       "22       22                                         Bumpy road\n",
       "23       23                                      Slippery road\n",
       "24       24                          Road narrows on the right\n",
       "25       25                                          Road work\n",
       "26       26                                    Traffic signals\n",
       "27       27                                        Pedestrians\n",
       "28       28                                  Children crossing\n",
       "29       29                                  Bicycles crossing\n",
       "30       30                                 Beware of ice/snow\n",
       "31       31                              Wild animals crossing\n",
       "32       32                End of all speed and passing limits\n",
       "33       33                                   Turn right ahead\n",
       "34       34                                    Turn left ahead\n",
       "35       35                                         Ahead only\n",
       "36       36                               Go straight or right\n",
       "37       37                                Go straight or left\n",
       "38       38                                         Keep right\n",
       "39       39                                          Keep left\n",
       "40       40                               Roundabout mandatory\n",
       "41       41                                  End of no passing\n",
       "42       42  End of no passing by vehicles over 3.5 metric ..."
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv('signnames.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True label of images\n",
      "Image 1 : Speed limit (30km/h)\n",
      "Image 2 : Road work\n",
      "Image 3 : Children crossing\n",
      "Image 4 : Right-of-way at the next intersection\n",
      "Image 5 : Go straight or left\n",
      "Result of Predict\n",
      "Image 1 : Speed limit (30km/h)\n",
      "Image 2 : Road work\n",
      "Image 3 : Children crossing\n",
      "Image 4 : Right-of-way at the next intersection\n",
      "Image 5 : Go straight or left\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",
    "label_web=[1,25,28,11,37]\n",
    "print('True label of images')\n",
    "for i in range(len(label_web)):\n",
    "    print('Image',i+1,':',df['SignName'][label_web[i]])\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    web_predict=sess.run(logits, feed_dict={x:img_list,keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    result = np.argmax(web_predict,1)\n",
    "    print('Result of Predict')\n",
    "    for i in range(len(result)):\n",
    "        print('Image',i+1,':',df['SignName'][result[i]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Analyze Performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "All prediction is correct.\n",
    "So accuracy is 1.0."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Output Top 5 Softmax Probabilities For Each Image Found on the Web"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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. `tk.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": 34,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TopKV2(values=array([  1.00000000e+00,   1.36987222e-09,   2.18211185e-10,\n",
      "         8.36052256e-13,   3.05288184e-13], dtype=float32), indices=array([1, 4, 2, 0, 5], dtype=int32))\n",
      "TopKV2(values=array([  1.00000000e+00,   6.56038717e-16,   6.68115228e-18,\n",
      "         1.08783375e-19,   6.38254241e-20], dtype=float32), indices=array([25, 26, 31, 22,  5], dtype=int32))\n",
      "TopKV2(values=array([  1.00000000e+00,   1.25605060e-18,   9.08736504e-25,\n",
      "         5.72260376e-28,   3.21337319e-28], dtype=float32), indices=array([28, 23, 20, 11, 29], dtype=int32))\n",
      "TopKV2(values=array([  1.00000000e+00,   1.40411083e-09,   3.75924777e-11,\n",
      "         1.43124609e-11,   3.17421497e-14], dtype=float32), indices=array([11, 30, 27, 28, 21], dtype=int32))\n",
      "TopKV2(values=array([  1.00000000e+00,   9.57622759e-10,   6.47817494e-11,\n",
      "         6.96238987e-12,   3.33516678e-12], dtype=float32), indices=array([37, 10, 39, 35, 31], dtype=int32))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAhoCAYAAADF+6hYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nB+uXI1pzNC9PAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb6a5bcef0>"
      ]
     },
     "metadata": {},
     "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",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    web_predict=sess.run(logits, feed_dict={x:img_list,keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    fig, ax = plt.subplots(len(web_predict),1,figsize=(10, 30))\n",
    "    plt.tight_layout()\n",
    "    for s in range(len(web_predict)):\n",
    "        softm=sess.run(tf.nn.softmax(web_predict[s]))\n",
    "        top5=sess.run(tf.nn.top_k(tf.constant(softm), k=5))\n",
    "        print(top5)\n",
    "        l=['a','a','a','a','a','a']\n",
    "        for i in range(5):\n",
    "            l[i+1]=df['SignName'][top5.indices[i]]\n",
    "            #print(l[i])\n",
    "        ax[s].set_xticklabels(l)\n",
    "        ax[s].bar(range(5),top5.values,align=\"center\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAABcCAYAAAB+6068AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5B0FBuzqYwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb34582860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,len(img_flist))\n",
    "img=np.zeros([32,32])\n",
    "img_list=np.zeros([len(img_flist),32,32,1])\n",
    "for im in range(len(img_flist)):\n",
    "    ax[im].axis('off')\n",
    "    img = preProcessImage(cv2.imread(img_flist[im]))\n",
    "    ax[im].imshow(img)\n",
    "    img_list[im] = np.reshape(img,(32,32,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "---\n",
    "\n",
    "## Step 4: 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": 36,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "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 maybe 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": 37,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAAEUCAYAAABanMWTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XmYXGWZ9/Hfnc6+r2SHsCRAZBPDMgICAgoODCAioL5E\nBkQdUfBVLtERxeUVhRFFBhnjQtAogsiqQWTfHCFBCQmQQMi+J2Qhe9Lp5/2jKtLpeu7TXd1VXecx\n38919ZX0XfWcpX5dp+qp5T4WQhAAAAAAID861HoDAAAAAAC7YqIGAAAAADnDRA0AAAAAcoaJGgAA\nAADkDBM1AAAAAMgZJmoAAAAAkDNM1AAAAAAgZ2o6UTOzeWa22cw2NPoZ1sZlnmBmiyq1jc46rjGz\nYGafa1K/oli/pkrrPcnMZprZJjN73Mz2qsZ6nHWTVcvX2dnM7ireZsHMTqj0OjLWTU4tX+fRZvaw\nma02s5Vm9jszG1rp9WSsn6xavs6xZjbVzNYUfx4xs7GVXk/G+smqdev/enE9J1dzPU3WSVYtX+eo\n4rIb31ZXV3o9Gesnq/LW293Mfmxmq8xsnZk9VY31OOsmq5av86NNbqdNxXW9qzXLy8M7ameEEHo2\n+llSy40xs44tvOprksY3qV1YrFecmQ2UdLekqyX1lzRV0h3VWFcGsmq5ZyR9TNKyKq7DQ04t00/S\nBEmjJO0lab2kW6u0Lg9ZtcwSSR9S4dg3UNL9kn5bpXV5yKoMZravCpktreZ6HGRVnr6NbqtvVXld\nTZFVy01Q4Rh4YPHfz1dxXTFk1QIhhF83vp0k/YekOZL+1prl5WGiFlV8tfsvZrbWzKZZo3clzOwi\nM3vVzNab2Rwz+2Sx3kPSg5KGNZ7xm9lEM/t2o/G7zOKLrxR8ycxekrTRzDoWx/3eCq+0z206G5c0\nRVJ3M3tHcRnvkNStWN+53H5m9ofiMtYU/z+i0eVPmNm1ZvZ88dWR+8ysv3OTfFDSyyGE34UQtki6\nRtKhZnZAK27eiiKrXYUQtoUQfhhCeEbSjlbfsBVGTrsKITxYvD+9FULYJOm/JR3T2tu3kshqVyGE\ntSGEeSGEIMlUuF/t18qbt6LIyvXfkr4kaVtZN2gVkVU6yKrk9thf0r9JujSEsDKEsCOE8EJrb99K\nIqtmjZf0y+LjV9lyOVEzs+GS/ijp2yq8avBFSb83s0HFq6yQdLqk3pIukvQDMzs8hLBR0mmSlrRi\nxn+BpH+V1FdSg6QHJE2TNFzSSZKuMLP3NxnzKxVm5VIxiCaXd1DhFfq9JO0pabMKD16NXSjp3yUN\nk1Qv6UfO9r2juD2SpOK+vlGs1wxZpYGcWuQ9kl5u4XWrhqx8ZrZW0hZJN0n6Tst2rXrIKs7MzpW0\nLYQwuYX7VHVklWm+mS0ys1ut8OmdmiKrqKMkzZf0DSt89HG6mZ3Twn2rGrLKZoWvKL0nsr4Wy8NE\n7d7iLHytmd1brH1M0uQQwuQQQkMI4WEVPur3AUkKIfwxhPBGKHhS0p8lHdfG7fhRCGFhCGGzpCMk\nDQohfLP47sgcST+VdH6TMZMkXWBmnYqXTWp8YQjhzRDC70MIm0II6yX9P0nHN1nGr0IIM4p/tFdL\n+rCZ1UW2r6ekdU1q6yT1asW+thZZtSyrWiOnMnMys0MkfU3Sla3c19YiqzKyCiH0ldRH0mWS/t76\n3W0VsmpBVmbWU4VJ9BVt3M+2IKuW3a9WFbdrL0nvUuH5xK/btMflI6uWZTVC0kEqPO8bpsIx8DYz\nO7BNe10esir/OeCFkp4OIcxt1Z5KaunnO6vprBDCI01qe0k618zOaFTrJOlxSTKz0yR9XdIYFSab\n3SVNb+N2LGyy/mHFV293qpP0dOMBIYQFZjZbhQel10MIC83sH5ebWXdJP5B0qgrfh5GkXmZWF0LY\n+ZG4xuudr8J+DpS0vMn2bVDhFYnGeqvwvZr2QlZvy8qq1sjpbc3mZGb7qfARjMtDCE/HrlNFZPW2\nFt2nQggbzex/JK00swNDCCvK2M+2IKu3ZWX1DRWe0LT6iUkFkNXb3KxCCBtUeFItScvN7DJJS82s\ndwjhrbL3tnXI6m1Z96vNkrZL+nYIoV7Sk2b2uKT3SXq13J1tJbJ6W0ufA16oNn76Iw8TtZiFKhzo\nP9H0AjPrIun3Kuz8fSGE7cWZ/c5bPPYZ0I0q/HHsNCRyncbjFkqaG0IY3YJt/aWkX6jwlm5TX5C0\nv6SjQgjLzOwwFV4FtkbXGdno/3uqcEdcFVnWy2r0ZUgrfL53X9X+o1pklQZyiih+LOERSd8KIfyq\nBdvWHsiqeTsf8Ier8NGaWiGrUidJGmFm/1H8fZCkO83seyGE77VgO6uFrJq3c3st81rVR1alXmrB\nttQCWTnM7BgV3v28qwXb5srDRx9jJkk6w8zeb2Z1ZtbVCl8oHCGps6QuklZKqi/O1t/XaOxySQPM\nrE+j2ouSPmBm/c1siJr/SMbzkt6ywhcWuxW34SAzOyJy3TuK678zclkvFV4FWWuFLx1+PXKdj1mh\n9XR3Sd+UdFej2Xtj90g6yMzOMbOuKnxM66UQwsxm9qXayCrCzLoUc5KkzsXbpZYPfuTUhBU+W/+Y\npJtDCP/TzPa3J7JqwsxOMbN3Frelt6QbJK1R+72S7CGrUiep8BGtw4o/SyR9UtLNzexLtZFVE2Z2\nlJntb2YdzGyACt+5eSKE0PRrFu2NrEo9JWmBpC9boYHGMZJOkPRQM/tSbWTlGy/p98WPUrZaLidq\nIYSFks6U9BUVAl6owndHOhR3+HMq3NBrJH1EhVbNO8fOlHS7pDlW+BztMBW+RDhN0jwVPh+b2da+\neMOfocKDzFwVZsw/U+G7EU2vuzmE8Ejxs7JN/VCFzjKrJP1V0p8i1/mVpIkqtHHvWty32DatlHSO\nCp+bXaPCF0ubfga33ZGVa5YKd/rhKhxIN6vwFn1NkFPUJZL2kfR1a3TOk6z9aA9kFdW3uF/rVGii\ntJ+kU0OhA27NkFV0m94MISzb+aNCh841xY/Z1QxZRe1THL9e0gxJW1Vo1FBTZBXdpu0q3CYfUOE4\n+FNJF9b6xXqyirPCC/UflnRb1va3hIXWdYtEBZjZE5ImhRB+VuttQTaySgM5pYOs0kFW6SCrdJBV\nOmqZVS7fUQMAAACA3RkTNQAAAADIGT76CAAAAAA5wztqAAAAAJAzbZqomdmpZjbLzGab2VWV2ihU\nHlmlg6zSQVbpIKs0kFM6yCodZJWuVn/00czqJL0m6RRJiyRNkXRBCOEVb0znzp1Dt27dSuo9evRw\n17N8efyE3507d3bHbNkS79ictZ7Nm2PdOqURI0ZE6x06lD/HXblypXvZxo0bo3XLOPWWl10IYZdB\nrcnKzPhMbDuoVVbe/WfQoEHumKVLl0brDQ0N5a4+SW3NqlevXmHAgAElde/YI0krVsTP5dypUyd3\nzPbt26P1jh07umO85Xl/J/3793eXtWFDvAt71vGvkprmJJWfVV1dXYjdXtu2bavotnqPI127do3W\nJWnHjvhpe/r27RutZz2GLFu2LFrPeqzs2bNntJ71mLzHHnuU1ObNm6dVq1bxWJWIf4bnFb169XIv\n855PecezLF26dInWs+6L3vPW1vhnyCrrGOQ9X/aOp5V+juIdA7PmU927dy+prV+/Xps3b272/Lr+\nI3fzjpQ0O4QwR5LM7LcqnEvBDb5bt246+uijS+qx2k433HBDtO5NoCRp5sz4aSUOO+wwd8y0adOi\n9S996UvRuheU5P9R/M//+OfUfe6556L1rAfs2B3b+UMpOyvUTLtkNWTIkGj9U5/6lDvmu9/9brT+\n1ltvVWSbElRWVgMGDNBXv/rVkvorr/jR3njjjdH6wIED3THehNp7Ii/5x9O99oqf+u+8885zl/Xs\ns89G6zffXNPzHZeVVceOHTVs2LCS+rx58yq6Ud7x/cADD3THrF27Nlr/4Ac/GK3X1dW5y/Lu04ce\neqg75thjj43WR44c6Y657LLLSmrjxo2LXZXHqnQkl9VRRx3lXlZfXx+tP/HEE2Wvx7svZL1Y5j1v\nrZDksjrooIPcy6ZMmRKte8fTTZs2VWSbdnKOXZmT7diYO+7IPEXcP7Tlo4/DVTix3U6LijXkD1ml\ng6zSQVbpIKs0kFM6yCodZJWwtryjFnu7ruTtHDO7VNKlUva7Q6iqsrNCzZBVOprNqnFOWR8XRNWV\nlVXWu1CoKo5/6SCrdJBVwtryjtoiSY3f3x0haUnTK4UQJoQQxoUQxmV9hh1VVXZW7bZlaIqs0tFs\nVo1zyvp+BKqurKyYqNUMx790kFU6yCphbXlHbYqk0Wa2t6TFks6X9JGsAQ0NDdHPimZ9wdL78nbv\n3r3dMbEvLkvZjRK872j86U9/itYfeOABd1n33ntvtH7ttde6Y84555xoPetLkO9///tLas7nqcvO\nCjXTLll5TQT+8Ic/uGN24++iecrOqtwmRJX8EvTq1avdy2JNTiRp0aJF0foXv/hFd1nf/OY3o/Uf\n/vCH7pgrrrjCvaxCysoqhOA27ShX1qTPW0fW9xbHjBkTrd9zzz3R+i233OIuy/ve4lVX+U3hjjzy\nyGj9mWeeccccf/zxJTWniQ6PVenIbVYnnXRStH7EEUe4Y7Ken5Vr9uzZZY/59a9/Ha1/9KMfbevm\nSDnOypPVZMt7XKz0d9E83ne0s/6GPvGJT5TUHnvssRatr9UTtRBCvZldJukhSXWSfhFCeLm1y0P1\nkFU6yCodZJUOskoDOaWDrNJBVmlryztqCiFMljS5QtuCKiKrdJBVOsgqHWSVBnJKB1mlg6zS1aYT\nXgMAAAAAKo+JGgAAAADkDBM1AAAAAMgZJmoAAAAAkDNtaiZSLjOLnvQ660TYhxxySLQ+d+5cd4zX\nunPVqlXumNGjR0frIZScE1CS30pVks4666xoPasV5ymnnFL2mKzTGgBZvNNe/OUvf2nnLdl9mFm0\nVXtWG+KRI0dG6wsXLix7/V5bdUn661//Gq1/+ctfjtanTp3qLuu1116L1t/1rndlbF2cd6oVSVqx\nYkXZy2upEIK2b99ekWVlnT/Pa/P86quvumO81uO//OUvo/WJEye6y/rsZz8brV9++eXuGO80DwMH\nDnTHVPI0E0BrVbIFf6U999xztd6EmujXr1+0/tJLL1VsHV/4whfcy7zHvmeffdYd4z1/WrBggTtm\nw4YNJbWWngKGd9QAAAAAIGeYqAEAAABAzjBRAwAAAICcYaIGAAAAADnDRA0AAAAAcqbduz526tQp\nWvf0798/Wp8xY4Y7pkuXLtF6VtfHPn36ROuvv/56tL7nnnu6y/K6m2V1HBoyZEi0/n/+z/9xxzz+\n+OMlta1bt7rXB5A/WcelHj16VGw9Wd0lPb17947Wr7zySnfM+973vmg9q4OjJ+s4W82uj95jVRav\ne1n37t3dMV5XYa8LcZZRo0ZF61kdis8999xo/b3vfa875uGHH47Wt2zZ4o6J/e3RCRLV8uijj5Y9\nplu3btF6a46be+21V7Q+f/58d8yPfvSjstfzz8B7jr9mzRp3jHesmzdvXrT+/e9/313WrbfeGq1n\ndX3M6orrWblyZUmtvr6+RWN5Rw0AAAAAcoaJGgAAAADkDBM1AAAAAMgZJmoAAAAAkDNM1AAAAAAg\nZ5ioAQAAAEDOtGt7/hBCtIVvVltfrwVz1hivHXLnzp3dMVOnTo3W169fH617pwCQpJ/+9KfR+skn\nn+yO8VqUXnDBBe6Y2G3gtXsGGvvEJz4RrXt/u6iM2P1z0KBB7vWzjnPlmjZtWtljbrzxxmj9vvvu\nc8d4Ld9/97vfuWP222+/aH3o0KEZW1c9HTp0yHy8iPHaSb/jHe9wx8yZMydaP/74490x3mPiEUcc\nEa3vvffe7rJuu+22aP0//uM/3DE9e/aM1o866ih3zLJly0pq27dvd68PtEXfvn2j9bVr17pjWtOG\n3+OdqiOrPX9rxI5Rqd2vYm3rm+O14W+Nrl27Ruuf/vSn3TG33HJL2et57rnnSmobN25s0VjeUQMA\nAACAnGGiBgAAAAA5w0QNAAAAAHKGiRoAAAAA5AwTNQAAAADImTZ1fTSzeZLWS9ohqT6EMC7r+vX1\n9dHOWHV1de6YDRs2ROs9evRwx/Tu3Tta9zrxSNKMGTOi9X/5l3+J1rO6cv35z3+O1keOHOmO8Tqy\nTZ482R2zbt26ktqOHTui1y03K9ROe2RVbkc7xJWTVQhB9fX1JfXRo0e7y//Zz37W9o1sgQ4d4q/Z\nHXnkkdH6woUL3WV99KMfjdazuj5u3bo1Wn/ppZfcMeUqJyszU7du3cpavncbZnWa69OnT7Teq1cv\nd4zXJc1bz5lnnuku61Of+lS0fvnll7tjvC6WixcvdscccsghJTUzc6/P41UaapnTRRdd5F42adKk\naD2rW7d3DGqNF198MVr3OpJL0qZNm8pez7Zt21p83VpmdcABB7iXzZw5M1r3OqFL0urVq9u8TTt5\nndVvvvnmiq1Dkp588smSmtdVvqlKtOc/MYSwqgLLQfWRVTrIKh1klQ6ySgdZpYGc0kFWCeKjjwAA\nAACQM22dqAVJfzazF8zs0kpsEKqGrNJBVukgq3SQVTrIKg3klA6ySlRbP/p4TAhhiZntIelhM5sZ\nQniq8RWKfxCXSlKnTp3auDq0QVlZoabIKh2ZWTXOKesz92gXLc6Kx6qaa3FWqCkeq9JBVolq0ztq\nIYQlxX9XSLpHUsk3z0MIE0II40II4zp2rMRX4tAa5WbV3tuHt5FVOprLqnFOWQ0iUH3lZJXV4ArV\nV05Wtdg+FPBYlQ6ySlerJ2pm1sPMeu38v6T3SYq3TkRNkVU6yCodZJUOskoHWaWBnNJBVmlry1tc\ngyXdU2yx21HSb0IIf8oasH37di1btqyk7rXgl/xW+7Nnz3bHjBsXfzHg2Wefdcc0NDRE6127do3W\n999/f3dZr7zySrT+mc98xh1z6aXxd5uzWjuPHTu2pPb888/Hrlp2VqiZdslq+fLllV7k7qisrLZt\n2xZtaz9o0CB3Bd5xKcvQoUOj9aVLl5a9nkWLFkXrH/rQh9xlffe7343Ws05P4rX7HzBggDumTGXf\nr0IIZa2gZ8+e0fr27dvdMd5t4rWslqSjjjoqWj/uuOOi9YsvvthdlncKgrvvvtsdc8wxx0Tr8+fP\nd8d4Lf0dPF6loV1y8u5XWR8l9+5zw4YNc8csWbKkvA1rhda04K+Qmt6nDjzwQPcy71hXyRb8rZE1\nx/BOW+M8/5YkDR48uKTW0o/Yt3qiFkKYI+nQ1o5H+yGrdJBVOsgqHWSVDrJKAzmlg6zSRnt+AAAA\nAMgZJmoAAAAAkDNM1AAAAAAgZ5ioAQAAAEDOtPuJzWKdtN566y33+h06xOeSWd1S9tlnn2j9pZde\ncsdcd9110fp//dd/Revr1q1zl/XLX/4yWv/Od77jjvG6f2V10jrssMNKapz/By1x11131XoTdkux\n7opZnV137NgRrWedk62+vj5aP+mkk9wxY8aMidZvueUWd4zH66w2adIkd8zxxx8frWd1inzzzTfL\n27AydOjQwe346/E6m82ZM8cd43X8vP32290xDzzwQLT+ox/9KFo/+eST3WV5j73XX3+9O+Zf/uVf\novWszm4rV64sqWV1wwR2uvLKK6P1rL+ffv36Reut6ex41llnRev33ntv2cvK4j3Xzer8Gzs3sXf8\nbw+f/exno/XHH3/cHdOlS5dofevWrWWv/8tf/nK0fu2115a9rB/84AfuZRMmTIjWs7o+Dhw4sKTW\n0nNL844aAAAAAOQMEzUAAAAAyBkmagAAAACQM0zUAAAAACBnmKgBAAAAQM60a9fHhoYGrV+/vqT+\nzne+0x2zZs2aaN3r6iNJS5cujdZHjx7tjtmyZUu03qNHj2g9qxvjs88+G60PHjzYHTN58uRo/eCD\nD3bHxLoEmZl7fQC1s23bNi1cuLCknnVc8MSOo81dduqpp7pjNm7cWPY2eH784x9H6/fcc4875tOf\n/nS0ntX9MHasi3UVbo2OHTtqjz32KGvMrFmzovWLLrrIHeN17zznnHPcMS+88EK0/qc//SlaHzp0\nqLssr7Pl+PHj3TFPPfVUtP7e977XHfPzn/+8pOY9tgONDR8+PFq/5JJL3DHldmzNcvjhh0frle76\nmNXd0RPbz02bNlVic1plwYIF0fqMGTPaZf1ZHZQryevQmdUt95lnnimptbTzLe+oAQAAAEDOMFED\nAAAAgJxhogYAAAAAOcNEDQAAAAByhokaAAAAAOQMEzUAAAAAyBmrVDvjFq3MLMTaWn7+8593x0yf\nPj1a37p1qztmypQp0XrWaQD23HPPstaf1W60Y8f4WQ8GDhzojvHamo4bN84dc/TRR5fU7r77bq1c\nubLNPfrNrP3+MHZjIQSySkRbs6qrqwvdu3cvqY8aNcod88orr0TrWa2cTznllGg9qxXwvHnzyqrv\nu+++7rK806Bce+217pgxY8ZE697pUSTp1ltvLal94xvf0Ny5c9t8nxo4cGA4/fTTS+r9+/d3xyxZ\nsiRaP/vss90xDz30ULS+ePFid8yAAQOide90AkOGDHGX5Z2W4bzzznPHeKeMyTplTewUFO9+97v1\nwgsvcPxLRK0eq7zTlyxfvrytm5OEiy++2L0sdtoLKX/PK/r06eNetm7dukqtxj2tyG233VaxdWTJ\nmk95p85qSVa8owYAAAAAOcNEDQAAAAByhokaAAAAAOQMEzUAAAAAyBkmagAAAACQM/H2hI2Y2S8k\nnS5pRQjhoGKtv6Q7JI2SNE/Sh0MIa5pdWceO6tevX0l94sSJ7hivI9rmzZvdMZs2bSprWZJ0/vnn\nR+uPPPJItJ7V9dHryrVs2TJ3TKdOnaL1Z5991h3Tq1evaN3MZrQ1K7QPskpHW7NqaGjQhg0bSupZ\nx5Ju3bpF61nHv1hnXUnasmWLO+a0006L1m+55ZZo/eqrr3aX9fTTT0frWV13vW5ZN910kzvmK1/5\nSklt2bJlMrMVauPjVV1dXbTDo3fbSn5X37Vr17pjvMuOO+44d8ztt98erXvdNh944AF3WYcccki0\nfuGFF7pjVq9eHa1ndW++7rrrSmpz586tSFZoH7V6rNpdujt6Zs+eXfaYvD2vqGRnxywnnnhitN6a\nro+xruo7/fWvf43WH3/8cXdMbNumTp3aom1pyTtqEyWd2qR2laRHQwijJT1a/B35RFbpIKt0kFUO\nFSdSPF4loDgRJqs0kVM6yCpxzU7UQghPSWr6EtqZknZOUW+TdFaFtwuVQ1bpIKt0kFUOFSdqPF4l\noGfPnhJZpYqc0kFWiWvtd9QGhxCWSlLx3/iZNpEHZJUOskoHWaWDrNJBVmkgp3SQVeKa/Y5aW5nZ\npZIulbI/34/aa5wV8o2s0kBO6Wiclff9X+QD96t0kFU6yCqfWjtzWm5mQyWp+O8K74ohhAkhhHEh\nhHFM1GqiVVm129ahMbJKR4uyIqdcKDsrr4kLqo77VRp4rEoHWSWutTOn+yWNL/5/vKT7KrM5qAKy\nSgdZpYOs0kFW6SCrNJBTOsgqcS1pz3+7pBMkDTSzRZK+Lum7ku40s4slLZB0bktWZmbq2LF0lUuX\nLnXHeC2HFy1a5I456aSTovUVK9wXEvTYY49F63vttVe0ftFFF7nLOvzww6P1ESNGuGPMLFpfvHix\nO2bcuF1f9JgzZ87O1sn7tzUrtBuyaiGvhbt336mCds8qqw2/Z8qUKdG611Zdkv71X/+1rHXMnTvX\nvWzy5MnRet++fd0xn/3sZ6P1D3/4w+6YjMeN/1UbH6/q6uqiH3/cvn27O8Y7vk+bNs0d07Vr12jd\na/UvSfX19dH6X/7yl2h9+PDh7rLuvffesuqS9Itf/CJaz2qBfeyxx5bUXn/9dakCWaHd8FjVQtde\ne220ftVVfsNF73HsySefdMccf/zxu/z+yiuvaNWqVQoh7JZZffzjH6/YsrxT1kj+Y88HPvABd8xD\nDz1UUrv00pZ9yrTZiVoI4QLnovhsCDWzzz77aMuWLdq4cWPTk7KRVU6FEMgqEWSVhhDC0EiZrHJm\n0KBBmjdvHlklguNf/o0dO1YvvPCC1q9fT1b/RPjSGAAAAADkDBM1AAAAAMgZJmoAAAAAkDNM1AAA\nAAAgZ8zrpFYNgwYNCmeddVZJfdKkSe6YUaNGReszZ850xzTthrjT0KGx7y0XxLpRStJ73vOesq4v\n+d3Nli9f7o754x//GK1/6UtfcsfEusj8+te/1vLly9vcBs/M2u8PYzcWQiCrRLQ1q7q6uhDr8rdp\n06a2LLbEkUceGa1ndR/83e9+F617xyWvG65U6D4bs3DhQndMz549o/UvfOEL7ph3v/vd0Xol7lM9\ne/YMhx56aEm9e/fuZW9P1uPOySefHK1Pnz7dHdO009tOS5YsidazHncGDx4crXfp0sUds2PHjmh9\n27Zt7phYF7yHHnpIq1ev5viXCB6r0kFW+RN7zFy6dKm2bt3abFa8owYAAAAAOcNEDQAAAAByhoka\nAAAAAOQMEzUAAAAAyBkmagAAAACQM0zUAAAAACBn2rU9f58+fcLRRx9dUs9qG92rV69ofcGCBe6Y\nrDbBlXLCCSe4ly1dujRaHzJkiDtm/vz50frBBx/sjunRo0dJjZbHaaGNbjqq1Z6/Qwf/9TLv+Lxx\n48a2bEqbZbWpb83pBrzTnfTp08cds2HDhpLatm3b1NDQ0Ob7lHcqmdg6d/Iek8aOHeuO8Vrtv/76\n6+4Y77b36rHTuOzU0NBQVj3rss2bN7tjYqcOWLRoUYtaUzeH41/74LEqHWSVjpZkxTtqAAAAAJAz\nTNQAAAAAIGeYqAEAAABAzjBRAwAAAICcYaIGAAAAADnTrl0fzWylpJ3tDQdKWtVuK8+fau3/XiGE\nQW1dCFn9QzX3nawqK9dZNclJIiuOf2kgq3SQVTrIKh01zapdJ2q7rNhsaghhXE1WngMp7X9K21pp\nqe17attbSante2rbW0kp7XtK21oNKe1/SttaDSntf0rbWg0p7X9K21oNtd5/PvoIAAAAADnDRA0A\nAAAAcqZdTvGuAAAgAElEQVSWE7UJNVx3HqS0/ylta6Wltu+pbW8lpbbvqW1vJaW07yltazWktP8p\nbWs1pLT/KW1rNaS0/yltazXUdP9r9h01AAAAAEAcH30EAAAAgJypyUTNzE41s1lmNtvMrqrFNrQX\nM/uFma0wsxmNav3N7GEze734b79abqNnd8pJIquUkFU6yCodZJUOskpHqlmRUz5yaveJmpnVSbpZ\n0mmSxkq6wMzGtvd2tKOJkk5tUrtK0qMhhNGSHi3+niu7YU4SWaVkosgqFRNFVqmYKLJKxUSRVSom\nKrGsyOkfap5TLd5RO1LS7BDCnBDCNkm/lXRmDbajXYQQnpK0ukn5TEm3Ff9/m6Sz2nWjWma3ykki\nq5SQVTrIKh1klQ6ySkeiWZFTQc1zqsVEbbikhY1+X1Ss7U4GhxCWSlLx3z1qvD0x5FRAVukgq3SQ\nVTrIKh1klY68Z0VOBTXPqRYTNYvUaD2ZP+SUDrJKB1mlg6zSQVbpIKs0kFNO1GKitkjSyEa/j5C0\npAbbUUvLzWyoJBX/XVHj7YkhpwKySgdZpYOs0kFW6SCrdOQ9K3IqqHlOtZioTZE02sz2NrPOks6X\ndH8NtqOW7pc0vvj/8ZLuq+G2eMipgKzSQVbpIKt0kFU6yCodec+KnApqn1MIod1/JH1A0muS3pD0\nn7XYhnbc19slLZW0XYVXKC6WNECF7jGvF//tX+vt3N1zIqu0fsgqnR+ySueHrNL5Iat0flLNipzy\nkZMVNw4AAAAAkBM1OeE1AAAAAMDHRA0AAAAAcoaJGgAAAADkDBM1AAAAAMgZJmoAAAAAkDNM1AAA\nAAAgZ5ioAQAAAEDOMFEDAAAAgJxhogYAAAAAOcNEDQAAAAByhokaAAAAAOQMEzUAAAAAyBkmagAA\nAACQM0zUAAAAACBnmKgBAAAAQM4wUQMAAACAnGGiBgAAAAA5w0QNAAAAAHKGiRoAAAAA5AwTNQAA\nAADImZpO1MxsnpltNrMNjX6GtXGZJ5jZokpto7OOa8wsmNnnmtSvKNavqdJ6P2xmr5rZejN7xczO\nqsZ6nHWTVXnrvcTMZhdvpz+19bZqZl1k0/J1djazu4q3WTCzE5pcbmb2PTN7s/hznZlZhdZNTi1f\nZ3M5nWhmj5vZOjObV4X1k1XL19lcVlea2Yzi49ZcM7uywusnq5avs7msrjCzOWb2lpktMbMfmFnH\nCq6frFq+zsysmlxvZqVvA7Iqa53N3a+uMbPtTW7LfVq6/Dy8o3ZGCKFno58ltdyYMg5Kr0ka36R2\nYbFecWY2XNIkSf9XUm9JV0r6jZntUY31OciqBczseEnfkXSmpP6S5kq6vRrraoRsWu4ZSR+TtCxy\n2aWSzpJ0qKRDJJ0u6ZMVXDc5tVxWThsl/UKF42C1kFXLZWVlxfX3k3SqpMvM7PwKr5+sWi4rqwck\nHR5C6C3pIBWOg5+LXK8tyKrlsrLa6UpJK6q0frJqueayuqPJbTmnpQvOw0QtysyONrO/mNlaM5vW\neIZqZhfZ2+8szTGzTxbrPSQ9KGlY41cAzGyimX270fhdZvXFWfCXzOwlSRvNrGNx3O/NbGXxVcCm\nB6spkrqb2TuKy3iHpG7F+s7l9jOzPxSXsab4/xGNLn/CzK41s+et8MrwfWbW37lJRkhaG0J4MBT8\nUYUnK/u25vatJLIqcYak34UQXg4hbJP0LUnvMbN2z4psdhVC2BZC+GEI4RlJOyJXGS/p+yGERSGE\nxZK+L+njLbmt24KcdtVcTiGE50MIv5LU4ge7SiGrXbUgq+tCCH8LIdSHEGZJuk/SMS2+wduArHbV\ngqzeCCGs3bloSQ2S9mvJbd1WZLWrFjxWycz2VmFycG2LbuQKIatdtSSrtsjlRM0K7x79UdK3VXhH\n4ouSfm9mg4pXWaHCK929JV0k6QdmdngIYaOk0yQtacUrABdI+ldJfVU4OD0gaZqk4ZJOknSFmb2/\nyZhfqTBLlwpP8H7Z5PIOkm6VtJekPSVtlvTfTa5zoaR/lzRMUr2kHznbN1XSq2b2b2ZWZ4WPPW6V\n9FIL968qyCrKij+Nf5cKr1C2G7JplXcUt3enacVa1ZBTOsiqbczMJB0n6eW2LqsF6yKrVjCzj5jZ\nW5JWqfCO2k9au6wy1klWrXOTpK8U19MuyKrVzjCz1Wb2spl9upyBeZio3Vucla81s3uLtY9JmhxC\nmBxCaAghPKzCROUDkhRC+GPxlZ8QQnhS0p9VOPi3xY9CCAtDCJslHSFpUAjhm8WZ8hxJP5XU9OMa\nkyRdYGadipdNanxhCOHNEMLvQwibQgjrJf0/Scc3WcavQggzin/EV0v6sJnVNd24EMIOFf7QfqPC\nBO03kj5ZHNdeyKoFWUmaXLzsEDPrJulrkoKk7m3b7Uxk07JsmtNT0rpGv6+T1LP4BLMSyKkyObUH\nsqp8Vtfo7SdIlURWFcoqhPCbUPjo4xhJ/yNpeWuWk4GsKpCVmZ0tqWMI4Z5yx5aBrCpzv7pT0oGS\nBkn6hKSvmdkFLR1csS+JtsFZIYRHmtT2knSumZ3RqNZJ0uOSZGanSfq6CgeSDio8AZ7exu1Y2GT9\nw8xsbaNanaSnGw8IISwws9kqfB/p9RDCwsbP58ysu6QfqPC5/H7Fci8zqwuFiVfT9c5XYT8HqsnB\n0cxOlnSdpBMk/U3SuyTdb2anhRBeLH93W4Ws3uZmFUJ41My+Lun3kvoUl7teUjW/REs2b3OzaYEN\nKrwSuFNvSRtCCKHM5XjI6W1tyak9kNXb2pyVmV2mwivUx4UQtrZmGRnI6m0VuV+FEF43s5cl/VjS\nB1u7nAiyelursrLCxwivU3FyVEVk9bZW369CCK80+vUvZnajpA+phb0L8jBRi1mowkz2E00vMLMu\nKjwBvlDSfSGE7cWZ/s4EYk+oNmrXdzOGRK7TeNxCSXNDCKNbsK2/VOFL7RdFLvuCpP0lHRVCWGZm\nh0n6u3b9WNzIRv/fU9J2FT5y0NRhkp4KIUwt/j7FzJ6TdLKk9pqoxZBVbANDuFnSzZJkZmMkfVXS\njBZsYyWRTfleVuHjPs8Xfz9U1f+YFjmlg6xawcz+XdJVkt4TQqhq17dGyKrtOqp9vgdPVuUZLWmU\npKeLk4/OkvqY2TJJR4cQ5pW5vHKQVduFJuvJlIePPsZMUuHznO+3wvexulrhC4YjVPiD7CJppaT6\n4uz9fY3GLpc0wMz6NKq9KOkDZtbfzIZIuqKZ9T8v6S0rfIGxW3EbDjKzIyLXvaO4/jsjl/VS4XOv\na63wJcSvR67zMTMbW5zdf1PSXY1m841NkXRc8Y9JZvZOFd5Orul31ERWJVkVb4ODrGBPSRMk3RhC\nWNPMvlQa2USYWRcz61r8tXPxdtl50PylpP9rZsOt0Ir4C5ImNrOfbUVOEVk5mVmH4mWdCr9aVzPr\n3Mx+VgJZRTST1UdVeFX7lFBGp7MKIKuIZrK6xIqdpM1srKQvS3q0mf2sBLKKyMhqhgqTiMOKP5eo\ncDscpl3fBaoGsopo5n51phWal5iZHalCJ9X7mtnPf8jlRC2EsFCF1uZfUSHwhSq0IO1Q/Czp51S4\n4ddI+oik+xuNnanC24lzrPC52mEqfKlwmqR5Knxe9o5m1r9Dhc59h6nQWn2VpJ+p8DG2ptfdHEJ4\npPjZ2aZ+qEKnmVWS/irpT5Hr/EqFJ4LLJHWV0wq3+FnfayTdZWbrVXjV4jshhD9n7Uu1kVVUVxW+\nQ7hBhYPK/6rw+eZ2RTauWSocoIdLeqj4/72Kl/1EhS8qT1fhwfCPqvKX6cnJlZXTe4q/T9bbXwSv\n+rGQrFxZWX1b0gAVPgWys9vb/2TtZyWQlSsrq2MkTTezjSrctyarcPtVFVm5olmFQgfVZTt/JK2W\n1FD8veJdBxsjK1fW/ep8SbNV+ArMLyV9L4RwW9Z+NmahYl+9QLnM7AlJk0IIP6v1tiAbWeUX2aSB\nnNJBVukgq3SQVTrylFUu31EDAAAAgN0ZEzUAAAAAyBk++ggAAAAAOcM7agAAAACQM22aqJnZqWY2\ny8xmm9lVldooVB5ZpYOs0kFW6SCrNJBTOsgqHWSVrlZ/9NHM6iS9JukUSYtUOM/XBWHXM3A3HcPn\nLNtBCGGXE+lVMquOHf1zpHfoEJ/379jhd4vt1q1bWXVJ6tw5frqkLVu2ROsbN250l2UWP+dg1n72\n6NGjImOWLVumtWvXVi0rVFZb71edO3cO3bt3L6mvW7eu8htbpr59+0brmzZtita9+40kbd26tez1\nd+nSJVrv16+fO2bQoEEltUWLFmn16tUlG1duVu11n4r9PUjZx8z6+vpovXfv3tH6mjX+6Ru7du0a\nrXt5ZNm+fXtZ69m4caO2bNlStePfHnvs4W6P9/fbqVMnd8y2bduide82lPzHHm9ZdXV17rK8/cn6\nW1m/fn207j1WS/HHsdWrV2vjxo1Vyypre7xMssZ4mTQ0NLhjvNvR+1vZsGGDuyzv/tOrVy93jLee\nrG2O7eeaNWu0YcMGnlckounzihj/mWXzjpQ0e+cJLM3styqcW8ENHjVTsay8J3SS/6TDe7CQpHe8\n4x3R+sEHH+yOGTVqVLQ+c+bMaP1///d/3WV5k74BAwa4Y4466qhofeDAge6YI44oPRfjpZdeGrsq\n96t0lJVV9+7dddxxx5XU//CHP1RzG1vkxBNPjNanTZsWrWe9KPHaa6+Vvf6RI0dG6x/60IfcMZ/+\n9KdLaqeffrp39arfr7wnjllPtMaOHRutr1271h2zcuXKaP2UU06J1u+8M3au1wLvWLrffvu5YzxL\nlixxLzvwwANLag8++GDsqhXL6bzzznMv814IHDp0qDtmwYIF0fqYMWPcMc8//3xZy/Im25J02WWX\nRetvvfWWO+aJJ56I1rMml4MHDy6p3XDDDbGrViyrnj17upd5E9SsMQcccEC0vnlz7HRaBd59zjvW\nPf300+6yvPuVd5yV/OciWdscu199//vfj12V5xUJa8tHH4dr1zOgLyrWkD9klQ6ySgdZpYOs0kBO\n6SCrdJBVwtryjlrs7bqSt0rN7FJJ0bcO0G7IKh1klY5ms2qcU9bHeVF1ZWWFmuH4lw6ySgdZJawt\nE7VFkhp/XmWEpJLPPoQQJkiaIPGZ1xoiq3SQVTqazapxTn379iWn2ikrK+5TNcPxLx1klQ6ySlhb\nJmpTJI02s70lLZZ0vqSPVGSrUGkVyyrrC7xeY5r+/fu7Y7zvBGR9mX3u3LnRuvdZbu+7G5I0f/78\naP3+++93x3zwgx+M1rO+1zZkyJCSmvPZ9936ftWnTx/3sjw03Wjinyar1atXR+tz5syJ1t/97ne7\nyzrppJOi9VtuucUd4zXo2WeffdwxL7/8ckkt4/scFckqq9mD952ZrL9b71gWa5Syk3fMnDVrVrT+\n+c9/3l3WD37wg2g967vI3ndxX331VXfMscceW1Kr9vEvq6mN972jrO8Txr67JUnDhg1zx3jfj/7M\nZz4TrWd9N/Hyyy+P1rPyPfTQQ6P1v//97+6YWKMT57G9Yll538+S/O/Ted/VlPzvxWd9B9FrDuI1\nybnmmmvcZX3lK1+J1r2GZ5J0wgknROtZ37GPfUfN8U/zWLU7avVELYRQb2aXSXpIUp2kX4QQSh85\nUXNklQ6ySgdZpYOs0kBO6SCrdJBV2tryjppCCJMlTa7QtqCKyCodZJUOskoHWaWBnNJBVukgq3S1\n6YTXAAAAAIDKY6IGAAAAADnDRA0AAAAAcoaJGgAAAADkTJuaiWD3M2rUKPeyxYsXR+uHHHKIO8Zr\n9ZzFa717zDHHROuTJ/vfn/XaUHutkCVp48aN0brXYlySxowZU1Lz2g7vznLYgr9izMxtC15JXgvq\ncePGuWO8++5hhx0Wra9YscJd1ogRI6L10047zR3j3adip7XYKdYOPqv9dSXs2LHDvay+vj5az2p1\n77Ulnzlzpjtm3333LWvM9773PXdZ73nPe6L1s88+2x1z8sknR+ujR492x0yfPr2klnEqharzTqXQ\nqVMnd4yX/U9/+lN3jPc3MW3atGjda6cvSb17947WJ0yY4I658MILo/Xf/OY37pjrr7++pFbt41bs\nlAA7eW3zR44cGa1L0h577BGtZ512x7t933rrrWh9+PDh7rLuu+++aP3MM890x3inN/JOCyFJb7zx\nRkkt67QUSBPvqAEAAABAzjBRAwAAAICcYaIGAAAAADnDRA0AAAAAcoaJGgAAAADkDF0fUZY+ffq4\nl/3973+P1r3OZpLUpUuXaN3rQidJ559/frTudeV64okn3GV5HcxOPPFEd4zXmXDixInumHPPPbek\n1tDQ4F4fperq6qL1rE58eWNmVV+Hd5/q1q2bO8brrPbv//7v0XpW97TZs2dH61kdHPv37x+tr1mz\nxh0T69zXoUN1X3v0Oi5K8Q5skr9vkrTnnntG61kdEb31eLK6Pn71q18ta1mSNGXKlGg9q+tjrFNv\nte+3q1evdi/zHseyus5u2rQpWvce9yS/6+JNN90UrXv3Q8nv0OnlkXVZ9+7d3THDhg0rqWU9hlfC\nQQcd5F42Z86caN3r3CnF90HK7hS5ffv2aN3reDl+/Hh3WT//+c/dyzy33XZbtJ71nOuMM84oqXnH\nf6SLd9QAAAAAIGeYqAEAAABAzjBRAwAAAICcYaIGAAAAADnDRA0AAAAAcoaJGgAAAADkDO35UZbp\n06e7l/Xr1y9az2pJ3rVr12h927Zt7pgtW7ZE67fffnu0ntXq2mtT/NJLL7ljvLbAPXr0cMcsX768\npOa1A0ZcSm34PSGEqq9j4MCB0frGjRvdMSeccEK07t13jzvuOHdZ3v3wox/9qDvmqKOOita3bt3q\njokdO7xW2rU0duxY97KlS5dG696pKLLss88+0frjjz/ujvFOv3DllVe6Y66//vpoferUqe6YcePG\nuZdVS+yUADsNHjw4Wvda8EtSfX19tD5mzJjyNkz+qSqeeeYZd8zhhx8erWedMuJvf/tbtO7tiyTN\nnz+/pJZ1P6yErNMSdOrUKVrfb7/93DHefmedsmHQoEHRemtOqeIda3/84x+7Yy677LJoPeuUEbU4\n7QXaH++oAQAAAEDOMFEDAAAAgJxhogYAAAAAOcNEDQAAAAByhokaAAAAAORMm1pkmdk8Sesl7ZBU\nH0Jo/9ZOaJHWZBXrdpTVnal///7RelZ3pt69e0frH/rQh9wxXgeq5557LlrP6oo1bdq0aP2YY45x\nx6xduzZav/DCC90xzz77bEnNuy25X6WjnKxCCO3S9bFv377R+uzZs90xZ599drTudSbN6o7n3d+y\nuq96XdLuvPNOd8w555xTUmtoaHCvX4n7ldelVpJGjhwZrS9atMgd43UmzOrQ6Rk6dGi07nWTk6SJ\nEydG61nH34MOOihanzFjhjtm//33L6nNmTPHvX4lssp63PEMGzbMvSx2DJeyHytWrlwZrb/xxhvR\netbx4bHHHovWjzzySHeM14nz3/7t39wx5ajUY1VWV0mvQ+e8efPcMW+++Wa0PmLECHdMnz59ovWF\nCxdG6yeeeKK7rM997nPR+k033eSO8XL861//6o6JHSe8YyDPK9JViV7GJ4YQVlVgOag+skoHWaWD\nrNJBVukgqzSQUzrIKkF89BEAAAAAcqatE7Ug6c9m9oKZXVqJDULVkFU6yCodZJUOskoHWaWBnNJB\nVolq60cfjwkhLDGzPSQ9bGYzQwhPNb5C8Q+CP4raI6t0kFU6MrNqnFO3bt1qtY0oaHFWqDmySgOP\nVekgq0S16R21EMKS4r8rJN0jqeTbkCGECSGEcXxxsbbIKh1klY7msmqcU+fOnWuxiSgqJ6tabB/e\nRlZp4LEqHWSVrlZP1Mysh5n12vl/Se+T5Ld9Qs2QVTrIKh1klQ6ySgdZpYGc0kFWaWvLRx8HS7qn\n2MK9o6TfhBD+VJGtQqWVnVWHDh3Us2fPkvpbb73ljnnXu94VrXvtwiXphhtuiNavuOIKd8ykSZOi\n9TFjxkTrN954o7uscePiLxxltXZesWJFtD569Gh3zD777FNSi53+QLv5/SqrLfhdd93VjlvSImVn\nVV9fX5EVZ7WZ3rJlS7R++OGHu2O2bdsWra9ZsyZaX7Bggbss737w1FNPRetZvJb3Uvx0Axltvity\nv3r55Zfdy3r16hWtd+nSxR3jtdLeY4893DHesdE7lcLf/vY3d1mPPvpotP7BD37QHfOJT3wiWr/8\n8svdMXV1dSU15/gnVSirrFMceKcyyHp8806zEHuc3OknP/lJtO4d5w499FB3WXfccUe07j3uSdJp\np50WrS9ZssQdM2XKlJKaczqOij1WzZo1y73s6KOPjtazjqXeaRFGjRrljjn44IOj9R49ekTr06dP\nd5fl8Y6zkn+fe/75590xsWO9c0zZrZ9XpK7VE7UQwhxJ/lEFuUFW6SCrdJBVOsgqHWSVBnJKB1ml\njfb8AAAAAJAzTNQAAAAAIGeYqAEAAABAzjBRAwAAAICcaesJr/FPbMeOHWVd3+sqN3jwYHfMd77z\nnWh93rx57piXXnopWv/5z38erWd1qNt///2j9RdeeMEdc8wxx0TrU6dOdccce+yxJbVOnTq5199d\n5bCzY8WYmSp1LjWvw6okLVy4MFrfvHmzO+aBBx6I1hcvXhytjx8/3l2W123viSeecMcsX748Wvfu\nn5L0zDPPlNQ2bNjgXr9cHTuWPjxmdZqLdTZsbpu8v4esToLXX399tH7NNddE63fffbe7LM+LL77o\nXhY7lkn+/kvSnDlzSmoZHToromvXru5lXodMry753TuXLVvmjrnkkkui9SOOOCJav/POO91lzZgR\n76a+9957u2O8jpAPPvigO+a8884rqWXdltXmrdvrviz53Tu945nkP6/43ve+F62vW7fOXdaqVaui\n9Ztuuskd87WvfS1aHzRokDsmtj9Zf8NIE++oAQAAAEDOMFEDAAAAgJxhogYAAAAAOcNEDQAAAABy\nhokaAAAAAOQMXR8RVVdXpz59+pTUDz74YHfMiBEjovXVq1e7Yw477LBoPatD3Cc/+clo3eui+Oyz\nz7rL6tatW7Q+a9Ysd4zXrWzgwIHumIaGhpJaCMG9Pqprv/32cy+bPXt2VdbZ0NCgLVu2lDXG63iW\n1TF09OjR0brXrU+SvvGNb0Trhx56aLSe1RHW2+ZzzjnHHeN1u8vqfti/f/+SWqxTY2t07NhR/fr1\nK6mvXLnSHTNs2LBoPev453UMXLJkiTtm7dq10fott9wSrXvHOEnRfZSkn/70p+4YL8cJEya4Y269\n9Vb3smrxOh5K0vDhw6P1SZMmuWO84/tvf/tbd8zJJ58crb/vfe+L1hctWuQuyzN9+nT3sgMOOCBa\nP//8890xsU6kZlb2dlWKt27vfiBJvXv3jtZ79erljnnzzTej9ZkzZ0br27Ztc5flHQOzulnHnm9J\n0s033+yO+dWvflVS43nFPx/eUQMAAACAnGGiBgAAAAA5w0QNAAAAAHKGiRoAAAAA5AwTNQAAAADI\nGSZqAAAAAJAztOdHVIcOHdS9e/eS+t577+2O8VoLDx482B2zePHiaH3ixInumG9961vRutdSf/36\n9e6y5syZE62///3vd8e89NJL0foNN9zgjlmwYEFJLau9L6qrWi34s4QQopl7LfAl6eijj47WV6xY\n4Y5573vfG62/9dZb7pjY6SMk/zQAsdb4zW1b7Hiyk9dCPqvd/h133FFSGzdunHv9cpiZunTpUtaY\nV155JVrPOv559tlnH/eyv/3tb9G6d8oJr+235J9SYMiQIe6Yu+++O1ofP368O+aZZ55xL2sr77Hq\n/vvvd8c8/vjj0XrW/ao1x2vvceTvf/97tB57nGhOVst373QgZ599tjvmvPPOK3sbWsrLasOGDe6Y\np59+Olrfc8893TF1dXXRetZt5fFOA+CdBkWS6uvro/WDDjrIHeOddsh7jiJJjzzySEktdnoFpI13\n1AAAAAAgZ5ioAQAAAEDOMFEDAAAAgJxhogYAAAAAOcNEDQAAAAByptmuj2b2C0mnS1oRQjioWOsv\n6Q5JoyTNk/ThEMKa6m0mymFmM9qaVceOHTVw4MCSutcFTpKee+65aP3CCy90x8ydOzda/+1vf+uO\nmT59erTudVqaMGGCuyyv494JJ5zgjvE6dk2aNMkdc88990Trlchqd5fV6W/q1KllL6/p3/G3vvUt\nPfjgg23OqqGhQRs3biypjxw50h2zdevWaN3r5ib5HRmfeuopd8wll1wSrXfoEH8tb/jw4e6y/vzn\nP5c95otf/GK0fuutt7pjnnzyyZLa66+/LjNboTY+XtXX12vVqlVZV2mxrI52++67b7TudXbM4nVw\nXLZsmTvmgQceiNbnz5/vjhkwYEC0fvXVV7tj1qwpvbn33XffimTlPVZldTn1ujsef/zx7phbbrkl\nWj/22GPdMW+++Wa0Pnny5Gh98+bN7rIuvvjiaP2ss85yx3jdHU888UR3TNPHvpkzZ2r58uUVeawK\nIWjHjh0l9b59+7pjvK7NvXr1csd4x8DWPB7EtlfyM5T85yKvvvqqO8bMovWFCxe6Yz73uc/t8vt/\n/ud/avXq1Tyv+CfTknfUJko6tUntKkmPhhBGS3q0+DvyiazSQVY5dPrpp8cmU2SVQ71795Z4vEpC\n8ck0WeXckCFD1LVr16Zlcsqhs88+W3vttVfTMlklrtmJWgjhKUmrm5TPlHRb8f+3SfJf0kGtkVU6\nyCqH3vnOd8beWSKrHCqeo4nHqwT07NlTIqvc69u3b+zdHnLKoXHjxsXOIUdWiWvtd9QGhxCWSlLx\n3z0qt0moMLJKB1mlg6zSQVbpIKs0kFM6yCpxzX5Hra3M7FJJl1Z7PWi7xllxdvt8436VBu5T6eA+\nlY7GWUXeQUCONM7K+x4W8oFjYD619h215WY2VJKK/8a/mSsphDAhhDAuhOB/4x/V1KqsspqGoGq4\nX5tarMEAAAkDSURBVKWjRVlxn8qFsrPiCWXNlJ0VE7Wa4LEqHWSVuNZO1O6XNL74//GS7qvM5qAK\nyCodZJUOskoHWaWDrNJATukgq8S1pD3/7ZJOkDTQzBZJ+rqk70q608wulrRA0rnV3MjUhBCi9cce\ne8wdc9JJJ1VyE/Zva1YdOnSItr9taGhwxxx++OHRep8+fdwx3/nOd6L1rFb33/72t6P1Y445Jlrv\n16+fuyxv2+bNm+eO8Za3ePFid8xFF120y+9PPPGE5s2bpxBCm7P6Z3LyySe7lz3yyCPRelareu++\nmPWOyWmnnbbL7xs2bNC2bdukNt6vtm/fHm2VHunS9Q9r166N1t/97ne7Y7wW315bdUkaM2ZMtD5l\nypRoPavV/5133hmtZ90Pzz03fvMVuzhGTZw4saRWvN/+r9r4eNWlSxftt99+JfWsv7XZs2dH67FT\nMuy0bt265jalxKhRo6L1QYMGRetZ7fl79OgRrRf/3qO8+9RHPvIRd8ynPvWpktobb7whVSCrPn36\nlNxnJf80LpI0ePDgaN3bN8k/NY2XhyRdd9110br3eH/99de7y/La82fdr7zTwvzxj390xzTdtjvv\nvHPn332bH6u6du0aPdZ069bNHfPiiy9G6ytXrnTHLFmyJFrPOp54p3Po0qVLtO614M/y+9//3r3M\nO1XGEUcc4Y75yU9+ssvvN998s+bMmSNVICvkR7MTtRDCBc5FFZ1ZoHJCCE0fUcgqZ0444QTdf//9\nWrVqFVnlXM+ePbVu3TrV19eTVc716tVLq1atGhq5iKxyZtiwYZo5cyZZ5dyHP/xh3XLLLVq8eDHH\nv5z7zGc+o6997WuaM2cOWf0Tae1HHwEAAAAAVcJEDQAAAAByhokaAAAAAOQMEzUAAAAAyJmqn/B6\nd/TPcA6eDh06qGvXriX1rC5lY8eOjdZXrHBP26H9998/Wn/ttdfcMTNmzIjWjzvuuGj9s5/9rLus\nhQsXRusjRoxwx7z3ve+N1idMmOCOie1nVoex3ZXX2TGL9/cgte6+uHr16rLHtEQIIdpNz+s2Jknr\n16+P1rM6RW7YsCFaP/TQQ90x3v3wmWeeidaz7tMf//jHo/U99tjDHbPPPvtE61md3WJd0ryOl+Vq\naGiI3vZel0RJOuCAA6L1vn37umNeffXVaD2rC17HjvGH7auvvtodU64jjzzSvczrglfsNhd1yCGH\nlNS8v9Ny7dixI9odNWv5e++9d7Q+depUd8z27dujda/LqeR3SX744YfLqkvSlVde6V5WSR/72MdK\nalndTstRX1+vVatWldR79uzpjvE6ZL7rXe9yx3iZDBw40B3jHYe9DplZx+ChQ2M9cqR7773XHeMd\nUzdt2uSOiT1PqlRWyA/eUQMAAACAnGGiBgAAAAA5w0QNAAAAAHKGiRoAAAAA5AwTNQAAAADIGSZq\nAAAAAJAz1p4tws2MfuTtIITQ5vMD9O/fP5xyyikl9dmzZ7tjvHbIWa32DzvssGg9az1bt26N1r3T\nA0ybNs1d1hlnnBGtZ7UF91qmDxkyxB0Ta+v95JNPau3atW3Oqtb3q6yW5Z06dYrWY+20mzNq1Kho\nva6uzh0zfPjwaP2pp54qe/1tvV917949xNrgZ7WM9lozZ7WGnjt3btlj5s+fH60vW7YsWu/fv7+7\nrFdeeSVa79evnzumc+fO0XrWqQti+zNr1ixt2rSpJvcp71h29tlnu2MWLFgQrWcd/7zl3XfffdH6\n0qVL3WV5pxTw6lKhxXrM888/746J7eeSJUu0devWNmfVrVu3sN9++5XUZ86c6Y4ZNmxYtH7aaae5\nY7xTFtx1113uGK+F/MiRI8telrf+rOPfiy++GK0feOCB7pgBAwaU1G688UYtWrSozVnV1dWF7t27\nl9Q3btzojhk9enS07h3bJf8UElmnX9ixY0e03qdPn2h93rx57rK89vxZx+BZs2ZF62vWrHHH9OrV\nq6S2ceNG7dixI/nnFbuLljyv4B01AAAAAMgZJmoAAAAAkDNM1AAAAAAgZ5ioAQAAAEDOMFEDAAAA\ngJxp766PKyXtbC82UNKqdlt5/lRr//cKIQxq60LI6h+que9kVVm5zqpJThJZcfxLA1mlg6zSQVbp\nqGlW7TpR22XFZlNDCONqsvIcSGn/U9rWSktt31Pb3kpKbd9T295KSmnfU9rWakhp/1Pa1mpIaf9T\n2tZqSGn/U9rWaqj1/vPRRwAAAADIGSZqAAAAAJAztZyoTajhuvMgpf1PaVsrLbV9T217Kym1fU9t\neysppX1PaVurIaX9T2lbqyGl/U9pW6shpf1PaVuroab7X7PvqAEAAAAA4vjoIwAAAADkTE0mamZ2\nqpnNMrPZZnZVLbahvZjZL8xshZnNaFTrb2YPm9nrxX/71XIbPbtTThJZpYSs0kFW6SCrdJBVOlLN\nipzykVO7T9TMrE7SzZJOkzRW0gVmNra9t6MdTZR0apPaVZIeDSGMlvRo8fdc2Q1zksgqJRNFVqmY\nKLJKxUSRVSomiqxSMVGJZUVO/1DznGrxjtqRkmaHEOaEELZJ+q2kM2uwHe0ihPCUpNVNymdKuq34\n/9skndWuG9Uyu1VOElmlhKzSQVbpIKt0kFU6Es2KnApqnlMtJmrDJS1s9PuiYm13MjiEsFSSiv/u\nUePtiSGnArJKB1mlg6zSQVbpIKt05D0rciqoeU61mKhZpEbryfwhp3SQVTrIKh1klQ6ySgdZpYGc\ncqIWE7VFkkY2+n2EpCU12I5aWm5mQyWp+O+KGm9PDDkVkFU6yCodZJUOskoHWaUj71mRU0HNc6rF\nRG2KpNFmtreZdZZ0vqT7a7AdtXS/pPHF/4+XdF8Nt8VDTgVklQ6ySgdZpYOs0kFW6ch7VuRUUPuc\nQgjt/iPpA5Jek/SGpP+sxTa0477eLmmppO0qvEJxsaQBKnSPeb34b/9ab+funhNZpfVDVun8kFU6\nP2SVzg9ZpfOTalbklI+crLhxAAAAAICcqMkJrwEAAAAAPiZqAAAAAJAzTNQAAAAAIGeYqAEAAABA\nzjBRAwAAAICcYaIGAAAAADnDRA0AAAAAcoaJGgAAAADkzP8HiFitnsT0eukAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb344a7d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    act1=outputFeatureMap(np.reshape(img_list[1],(1,32,32,1)),conv2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAAEUCAYAAABanMWTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nADnz/wO+B+OKoRa0lwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb340eaeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    act1=outputFeatureMap(np.reshape(img_list[2],(1,32,32,1)),conv2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Question 9\n",
    "\n",
    "Discuss how you used the visual output of your trained network's feature maps to show that it had learned to look for interesting characteristics in traffic sign images\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "**Answer:**\n",
    "When the classifier makes an erroneous prediction, it can comprehend what kind of image is not enough for learning by comparing the feature map.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "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": {
    "deletable": true,
    "editable": true
   },
   "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": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 1. Provide a basic summary of the data set and identify where in your code the summary was done. In the code, the analysis should be done using python, numpy and/or pandas methods rather than hardcoding results manually.\n",
    "\n",
    "The code for this step is contained in the second code cell of the IPython notebook.\n",
    "\n",
    "I used the numpy to calculate summary statistics of the traffic signs data set:\n",
    "\n",
    "Number of training examples = 34799\n",
    "Number of testing examples = 12630\n",
    "Image data shape = (32, 32, 1)\n",
    "Number of classes = 43"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 2. Include an exploratory visualization of the dataset and identify where the code is in your code file.\n",
    "\n",
    "The code for this step is contained in the third code cell of the IPython notebook.\n",
    "\n",
    "Here is an exploratory visualization of the data set. It is a bar chart showing how the data ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Container object of 43 artists>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAD8CAYAAACRkhiPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYYaAJDXMEJCkhv1/jbAomGDjeJgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb2f745550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_label=np.zeros(len(dict_label))\n",
    "for v in y_train:\n",
    "    train_label[v]+=1\n",
    "valid_label=np.zeros(len(dict_label))\n",
    "for v in y_valid:\n",
    "    valid_label[v]+=1\n",
    "plt.bar(range(len(train_label)),train_label,alpha=0.3,color='b')\n",
    "plt.bar(range(len(valid_label)),valid_label,alpha=0.3,color='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Design and Test a Model Architecture"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 1. Describe how, and identify where in your code, you preprocessed the image data. What tecniques were chosen and why did you choose these techniques? Consider including images showing the output of each preprocessing technique. Pre-processing refers to techniques such as converting to grayscale, normalization, etc.\n",
    "\n",
    "The code for this step is contained in the 18th code cell of the IPython notebook.\n",
    "\n",
    "As a first step, I decided to convert the images to grayscale because the LeNet Architecture accept 1 channel image.\n",
    "\n",
    "Here is an example of a traffic sign image before and after grayscaling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb3401aa58>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAC7CAYAAAB1qmWGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb2c3a17b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2)\n",
    "ax[0].imshow(X_train[0])\n",
    "ax[1].imshow(cv2.cvtColor(X_train[0],cv2.COLOR_RGB2GRAY))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "As a last step, I normalized the image data because the images have different contrast.\n",
    "Here is an example of a traffic sign image before and after normalization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb2ebf1fd0>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAC7CAYAAAB1qmWGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyoB1zAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb2ec5e630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2)\n",
    "ax[0].imshow(cv2.cvtColor(X_train[0],cv2.COLOR_RGB2GRAY))\n",
    "ax[1].imshow(preProcessImage(X_train[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 2. Describe how, and identify where in your code, you set up training, validation and testing data. How much data was in each set? Explain what techniques were used to split the data into these sets. (OPTIONAL: As described in the \"Stand Out Suggestions\" part of the rubric, if you generated additional data for training, describe why you decided to generate additional data, how you generated the data, identify where in your code, and provide example images of the additional data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "My final training set had 42000 number of images. My validation set and test set had 4410 and 12630 number of images.\n",
    "\n",
    "The seventh code cell of the IPython notebook contains the code for augmenting the data set. I decided to generate additional data because the umber of examples per label is not equal.\n",
    "To add more data to the the data set, I used the affine transformation because it's same as changing camera angle.\n",
    "Here is an example of an original image and an augmented image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fdb2eaf3f60>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAC7CAYAAAB1qmWGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb34547b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2)\n",
    "ax[0].imshow(X_train[0])\n",
    "ax[1].imshow(generateFake(X_train[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 3. Describe, and identify where in your code, what your final model architecture looks like including model type, layers, layer sizes, connectivity, etc.) Consider including a diagram and/or table describing the final model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The code for my final model is located in the 25th cell of the ipython notebook.\n",
    "\n",
    "My final model consisted of the following layers:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "| Layer                    | Description | \n",
    "|:-------------------------|:------------|\n",
    "| Input                    | 32x32x1 image |\n",
    "| Convolution 3*3          | 1x1 stride, valid padding, outputs 30x30x6   |\n",
    "| RELU                     |  　　|\n",
    "| Convolution 3*3          | 1x1 stride, valid padding, outputs 28x28x16   |\n",
    "| RELU                     |  　　|\n",
    "| Max pooling              | 2x2 stride, outputs 14x14x16　|\n",
    "| Convolution 3*3          | 1x1 stride, valid padding, outputs 12x12x26   |\n",
    "| RELU                     |  　　|\n",
    "| Convolution 3*3          | 1x1 stride, valid padding, outputs 10x10x36   |\n",
    "| RELU                     |  　　|\n",
    "| Max pooling              | 2x2 stride, outputs 5x5x36　|\n",
    "| Convolution 5*5          |   1x1 stride, valid padding, outputs 1x1x4000   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 2000   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 1000   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 500   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 100   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 84   |\n",
    "| RELU                     |  　　|\n",
    "| Fully connected          | outputs 42   |"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 5. Describe the approach taken for finding a solution. Include in the discussion the results on the training, validation and test sets and where in the code these were calculated. Your approach may have been an iterative process, in which case, outline the steps you took to get to the final solution and why you chose those steps. Perhaps your solution involved an already well known implementation or architecture. In this case, discuss why you think the architecture is suitable for the current problem."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The code for calculating the accuracy of the model is located in the below cell of the Ipython notebook.\n",
    "\n",
    "My final model results were:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training set accuracy of  0.996666\n",
      "validation set accuracy of  0.972336\n",
      "test set accuracy of  0.947744\n"
     ]
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    training_accuracy = sess.run(accuracy_operation, feed_dict={x: X_train_prepro, y: y_train, keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    print('training set accuracy of ',training_accuracy)\n",
    "    validation_accuracy = sess.run(accuracy_operation, feed_dict={x: X_valid_prepro, y: y_valid, keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    print('validation set accuracy of ',validation_accuracy)\n",
    "    test_accuracy = sess.run(accuracy_operation, feed_dict={x: X_test_prepro, y: y_test, keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    print('test set accuracy of ',test_accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When trying LeNet for the first time, learning with noise added data did not work well.\n",
    "So I tried a deeper network, and the prediction result improved."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Test a Model on New Images"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 1. Choose five German traffic signs found on the web and provide them in the report. For each image, discuss what quality or qualities might be difficult to classify."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAABcCAYAAAB+6068AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5B0FBuzqYwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb2ea9cf98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Load the images and plot them here.\n",
    "### Feel free to use as many code cells as needed.\n",
    "img_flist=glob.glob('img/*')\n",
    "fig, ax = plt.subplots(1,len(img_flist))\n",
    "img=np.zeros([32,32])\n",
    "img_list=np.zeros([len(img_flist),32,32,1])\n",
    "for im in range(len(img_flist)):\n",
    "    ax[im].axis('off')\n",
    "    img = preProcessImage(cv2.imread(img_flist[im]))\n",
    "    ax[im].imshow(img)\n",
    "    img_list[im] = np.reshape(img,(32,32,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The first image has a part of the image in the shadow, so the brightness of the image is not uniform.\n",
    "In the second image, it is difficult to distinguish between a marker and a background because part of the background is red.\n",
    "The third and fourth images are not special.\n",
    "The fifth image is inclined."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 2. Discuss the model's predictions on these new traffic signs and compare the results to predicting on the test set. Identify where in your code predictions were made. At a minimum, discuss what the predictions were, the accuracy on these new predictions, and compare the accuracy to the accuracy on the test set (OPTIONAL: Discuss the results in more detail as described in the \"Stand Out Suggestions\" part of the rubric).\n",
    "\n",
    "The code for making predictions on my final model is located in the 33th cell of the Ipython notebook.  \n",
    "The model was able to correctly guess 5 of the 5 traffic signs, which gives an accuracy of 100%.  \n",
    "\n",
    "| Image                | Prediction | \n",
    "|:---------------------|:------------|\n",
    "| Speed limit (30km/h) | Speed limit (30km/h) |\n",
    "| Road work            | Road work            |\n",
    "| Children crossing    | Children crossing    |\n",
    "| Right-of-way at the next intersection |  Right-of-way at the next intersection |\n",
    "| Go straight or left  | Go straight or left |\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### 3. Describe how certain the model is when predicting on each of the five new images by looking at the softmax probabilities for each prediction and identify where in your code softmax probabilities were outputted. Provide the top 5 softmax probabilities for each image along with the sign type of each probability. (OPTIONAL: as described in the \"Stand Out Suggestions\" part of the rubric, visualizations can also be provided such as bar charts)\n",
    "\n",
    "The code for making predictions on my final model is located in the 11th cell of the Ipython notebook.\n",
    "I plot the the top 5 softmax probabilities for each image along with the sign type of each probability. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAWYCAYAAACxvtCBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdb2ea91748>"
      ]
     },
     "metadata": {},
     "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",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess,'./Deepnet')\n",
    "    web_predict=sess.run(logits, feed_dict={x:img_list,keep_prob1: 1.0,keep_prob2:1.0,keep_prob3:1.0,keep_prob4:1.0})\n",
    "    fig, ax = plt.subplots(len(web_predict),1,figsize=(10, 20))\n",
    "    plt.tight_layout()\n",
    "    for s in range(len(web_predict)):\n",
    "        softm=sess.run(tf.nn.softmax(web_predict[s]))\n",
    "        top5=sess.run(tf.nn.top_k(tf.constant(softm), k=5))\n",
    "        l=['a','a','a','a','a','a']\n",
    "        for i in range(5):\n",
    "            l[i+1]=df['SignName'][top5.indices[i]]\n",
    "            #print(l[i])\n",
    "        ax[s].set_xticklabels(l)\n",
    "        ax[s].bar(range(5),top5.values,align=\"center\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(Optional) Visualizing the Neural Network (See Step 4 of the Ipython notebook for more details)\n",
    "#### 1.Discuss the visual output of your trained network's feature maps. What characteristics did the neural network use to make classifications?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer:**\n",
    "When the classifier makes an erroneous prediction, it can comprehend what kind of image is not enough for learning by comparing the feature map."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}