Traffic_Light_Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Traffic Light Classifier\n",
    "---\n",
    "\n",
    "In this project, you’ll use your knowledge of computer vision techniques to build a classifier for images of traffic lights! You'll be given a dataset of traffic light images in which one of three lights is illuminated: red, yellow, or green.\n",
    "\n",
    "In this notebook, you'll pre-process these images, extract features that will help us distinguish the different types of images, and use those features to classify the traffic light images into three classes: red, yellow, or green. The tasks will be broken down into a few sections:\n",
    "\n",
    "1. **Loading and visualizing the data**. \n",
    "      The first step in any classification task is to be familiar with your data; you'll need to load in the images of traffic lights and visualize them!\n",
    "\n",
    "2. **Pre-processing**. \n",
    "    The input images and output labels need to be standardized. This way, you can analyze all the input images using the same classification pipeline, and you know what output to expect when you eventually classify a *new* image.\n",
    "    \n",
    "3. **Feature extraction**. \n",
    "    Next, you'll extract some features from each image that will help distinguish and eventually classify these images.\n",
    "   \n",
    "4. **Classification and visualizing error**. \n",
    "    Finally, you'll write one function that uses your features to classify *any* traffic light image. This function will take in an image and output a label. You'll also be given code to determine the accuracy of your classification model.    \n",
    "    \n",
    "5. **Evaluate your model**.\n",
    "    To pass this project, your classifier must be >90% accurate and never classify any red lights as green; it's likely that you'll need to improve the accuracy of your classifier by changing existing features or adding new features. I'd also encourage you to try to get as close to 100% accuracy as possible!\n",
    "    \n",
    "Here are some sample images from the dataset (from left to right: red, green, and yellow traffic lights):\n",
    "<img src=\"images/all_lights.png\" width=\"50%\" height=\"50%\">\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "### *Here's what you need to know to complete the project:*\n",
    "\n",
    "Some template code has already been provided for you, but you'll need to implement additional code steps to successfully complete this project. Any code that is required to pass this project is marked with **'(IMPLEMENTATION)'** in the header. There are also a couple of questions about your thoughts as you work through this project, which are marked with **'(QUESTION)'** in the header. Make sure to answer all questions and to check your work against the [project rubric](https://review.udacity.com/#!/rubrics/1213/view) to make sure you complete the necessary classification steps!\n",
    "\n",
    "Your project submission will be evaluated based on the code implementations you provide, and on two main classification criteria.\n",
    "Your complete traffic light classifier should have:\n",
    "1. **Greater than 90% accuracy**\n",
    "2. ***Never* classify red lights as green**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Loading and Visualizing the Traffic Light Dataset\n",
    "\n",
    "This traffic light dataset consists of 1484 number of color images in 3 categories - red, yellow, and green. As with most human-sourced data, the data is not evenly distributed among the types. There are:\n",
    "* 904 red traffic light images\n",
    "* 536 green traffic light images\n",
    "* 44 yellow traffic light images\n",
    "\n",
    "*Note: All images come from this [MIT self-driving car course](https://selfdrivingcars.mit.edu/) and are licensed under a [Creative Commons Attribution-ShareAlike 4.0 International License](https://creativecommons.org/licenses/by-sa/4.0/).*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import resources\n",
    "\n",
    "Before you get started on the project code, import the libraries and resources that you'll need."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import helpers # helper functions\n",
    "\n",
    "import random\n",
    "import numpy as np\n",
    "from scipy.optimize import curve_fit\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg # for loading in images\n",
    "import cv2 # computer vision library\n",
    "\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training and Testing Data\n",
    "\n",
    "All 1484 of the traffic light images are separated into training and testing datasets. \n",
    "\n",
    "* 80% of these images are training images, for you to use as you create a classifier.\n",
    "* 20% are test images, which will be used to test the accuracy of your classifier.\n",
    "* All images are pictures of 3-light traffic lights with one light illuminated.\n",
    "\n",
    "## Define the image directories\n",
    "\n",
    "First, we set some variables to keep track of some where our images are stored:\n",
    "\n",
    "    IMAGE_DIR_TRAINING: the directory where our training image data is stored\n",
    "    IMAGE_DIR_TEST: the directory where our test image data is stored"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Image data directories\n",
    "IMAGE_DIR_TRAINING = \"traffic_light_images/training/\"\n",
    "IMAGE_DIR_TEST = \"traffic_light_images/test/\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the datasets\n",
    "\n",
    "These first few lines of code will load the training traffic light images and store all of them in a variable, `IMAGE_LIST`. This list contains the images and their associated label (\"red\", \"yellow\", \"green\"). \n",
    "\n",
    "You are encouraged to take a look at the `load_dataset` function in the helpers.py file. This will give you a good idea about how lots of image files can be read in from a directory using the [glob library](https://pymotw.com/2/glob/). The `load_dataset` function takes in the name of an image directory and returns a list of images and their associated labels. \n",
    "\n",
    "For example, the first image-label pair in `IMAGE_LIST` can be accessed by index: \n",
    "``` IMAGE_LIST[0][:]```.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using the load_dataset function in helpers.py\n",
    "# Load training data\n",
    "IMAGE_LIST = helpers.load_dataset(IMAGE_DIR_TRAINING)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize the Data\n",
    "\n",
    "The first steps in analyzing any dataset are to 1. load the data and 2. look at the data. Seeing what it looks like will give you an idea of what to look for in the images, what kind of noise or inconsistencies you have to deal with, and so on. This will help you understand the image dataset, and **understanding a dataset is part of making predictions about the data**."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "### Visualize the input images\n",
    "\n",
    "Visualize and explore the image data! Write code to display an image in `IMAGE_LIST`:\n",
    "* Display the image\n",
    "* Print out the shape of the image \n",
    "* Print out its corresponding label\n",
    "\n",
    "See if you can display at least one of each type of traffic light image – red, green, and yellow — and look at their similarities and differences."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape: 40px X 70px\n",
      "Label: red\n",
      "Index: 331\n",
      "Shape: 35px X 91px\n",
      "Label: yellow\n",
      "Index: 734\n",
      "Shape: 18px X 41px\n",
      "Label: green\n",
      "Index: 766\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3e469198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## TODO: Write code to display an image in IMAGE_LIST (try finding a yellow traffic light!)\n",
    "## TODO: Print out 1. The shape of the image and 2. The image's label\n",
    "\n",
    "def getImage(index=0, color=None, Random=False, showImg=False):\n",
    "    #get total number of images in set\n",
    "    total_num_of_images = len(IMAGE_LIST)\n",
    "    # check whether color is red green of yellow\n",
    "    acceptable_colors = tuple(['red', 'yellow', 'green'])\n",
    "    if (not color in acceptable_colors) or color == None:\n",
    "            raise ValueError('color {} is not an acceptable color'.format(color))\n",
    "    # pick random index if random, and keep choosign random image until correct color is found\n",
    "    if Random == True:\n",
    "        image_index = random.randint(0, total_num_of_images-1)\n",
    "        if (color != None):\n",
    "            while IMAGE_LIST[image_index][1] != color:\n",
    "                image_index = random.randint(0, total_num_of_images-1)\n",
    "    # incement index until color is found\n",
    "    else:\n",
    "        image_index = index\n",
    "        while IMAGE_LIST[image_index][1] != color:\n",
    "            if image_index >= total_num_of_images:\n",
    "                total_num_of_images = 0\n",
    "            else:\n",
    "                total_num_of_images += 1\n",
    "    \n",
    "    # select image\n",
    "    selected_image = IMAGE_LIST[image_index][0]\n",
    "    # select label\n",
    "    selected_label = IMAGE_LIST[image_index][1]\n",
    "    \n",
    "    if showImg:\n",
    "        plt.imshow(selected_image)\n",
    "        print(\"Shape: {0[1]}px X {0[0]}px\".format(selected_image.shape))\n",
    "        print(\"Label: \" + str(selected_label))\n",
    "        print(\"Index: \" + str(image_index))\n",
    "    return selected_image, selected_label, image_index \n",
    "\n",
    "\n",
    "image, label, index = getImage(color='red', Random=True, showImg=True)\n",
    "image2, label2, index3 = getImage(color='yellow', Random=True, showImg=True)\n",
    "image3, label2, index3 = getImage(color='green', Random=True, showImg=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Pre-process the Data\n",
    "\n",
    "After loading in each image, you have to standardize the input and output!\n",
    "\n",
    "### Input\n",
    "\n",
    "This means that every input image should be in the same format, of the same size, and so on. We'll be creating features by performing the same analysis on every picture, and for a classification task like this, it's important that **similar images create similar features**! \n",
    "\n",
    "### Output\n",
    "\n",
    "We also need the output to be a label that is easy to read and easy to compare with other labels. It is good practice to convert categorical data like \"red\" and \"green\" to numerical data.\n",
    "\n",
    "A very common classification output is a 1D list that is the length of the number of classes - three in the case of red, yellow, and green lights - with the values 0 or 1 indicating which class a certain image is. For example, since we have three classes (red, yellow, and green), we can make a list with the order: [red value, yellow value, green value]. In general, order does not matter, we choose the order [red value, yellow value, green value] in this case to reflect the position of each light in descending vertical order.\n",
    "\n",
    "A red light should have the  label: [1, 0, 0]. Yellow should be: [0, 1, 0]. Green should be: [0, 0, 1]. These labels are called **one-hot encoded labels**.\n",
    "\n",
    "*(Note: one-hot encoding will be especially important when you work with [machine learning algorithms](https://machinelearningmastery.com/how-to-one-hot-encode-sequence-data-in-python/)).*\n",
    "\n",
    "<img src=\"images/processing_steps.png\" width=\"80%\" height=\"80%\">\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='task2'></a>\n",
    "### (IMPLEMENTATION): Standardize the input images\n",
    "\n",
    "* Resize each image to the desired input size: 32x32px.\n",
    "* (Optional) You may choose to crop, shift, or rotate the images in this step as well.\n",
    "\n",
    "It's very common to have square input sizes that can be rotated (and remain the same size), and analyzed in smaller, square patches. It's also important to make all your images the same size so that they can be sent through the same pipeline of classification steps!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "angle: 5.599339485168457\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c412940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# calculate the average of a lower bounded array of numbers\n",
    "# for classifying traffic light images this lower bound was \n",
    "# set to 0.1 radians.\n",
    "def average(array, lBound=0.1, uBound=0.6):\n",
    "    new_array = []\n",
    "    for i,row in enumerate(array):\n",
    "        for j,cell in enumerate(row):\n",
    "            if (cell >= lBound and cell < uBound):\n",
    "                new_array.append(cell)\n",
    "    if len(new_array) == 0:\n",
    "        return 0\n",
    "    else:\n",
    "        return sum(new_array)/len(new_array)\n",
    "    \n",
    "def to_polar_coords(xpix, ypix):\n",
    "    # Calculate distance to each pixel\n",
    "    dist = np.sqrt(xpix**2 + ypix**2)\n",
    "    # Calculate angle using arctangent function\n",
    "    angles = np.arctan2(ypix, xpix)\n",
    "    return dist, angles\n",
    "\n",
    "# Use a sobel image filter and return the magnitude and angles vectors\n",
    "# for the image\n",
    "# reference : https://www.programcreek.com/python/example/89325/cv2.Sobel\n",
    "def sobel_filter(image):\n",
    "    # sobel filter x and y gradients\n",
    "    sobelx = cv2.Sobel(image, cv2.CV_64F, 1, 0, ksize=5)\n",
    "    sobely = cv2.Sobel(image, cv2.CV_64F, 0, 1, ksize=5)\n",
    "    # absolute the gradients\n",
    "    abs_sobelx = np.absolute(sobelx)\n",
    "    abs_sobely = np.absolute(sobely)\n",
    "    \n",
    "    # get magnitude array & angle array\n",
    "    mag, ang = to_polar_coords(abs_sobelx, abs_sobely)\n",
    "    \n",
    "     # scale values to 8 bit\n",
    "    scale_factor = np.max(mag)/255\n",
    "    mag = (mag/scale_factor).astype(np.uint8)\n",
    "    \n",
    "    return mag, ang\n",
    "\n",
    "def getBoundingBox(img_thres):\n",
    "    # ref : https://www.pyimagesearch.com/2017/02/20/text-skew-correction-opencv-python/\n",
    "    # grab the (x, y) coordinates of all pixel values that\n",
    "    # are greater than zero, then use these coordinates to\n",
    "    # compute a rotated bounding box that contains all\n",
    "    # coordinates\n",
    "    coords = np.column_stack(np.where(img_thres > 0))\n",
    "    rect = cv2.minAreaRect(coords)\n",
    "    box = cv2.boxPoints(rect)\n",
    "    box = np.int0(box)\n",
    "    \n",
    "    angle = rect[-1]\n",
    "    # handle the angle to get correct output\n",
    "    if angle < -45:\n",
    "        angle = -(90 + angle)\n",
    "    else:\n",
    "        angle = -angle\n",
    "    \n",
    "    return box, angle\n",
    "            \n",
    "    \n",
    "def rectanglemask(img, box):\n",
    "    # transpose as box is flipped\n",
    "    mask_img = cv2.transpose(np.zeros_like(img))\n",
    "    cv2.drawContours(mask_img,[box], 0, (255,255,255), -1)\n",
    "    mask_img = np.uint8(mask_img)\n",
    "    ret, mask = cv2.threshold(mask_img,1,255,cv2.THRESH_BINARY)\n",
    "    # transpose back\n",
    "    mask = cv2.transpose(mask)\n",
    "    return mask\n",
    "\n",
    "# rotate the image by a specified angle.\n",
    "# ref: https://www.pyimagesearch.com/2014/01/20/basic-image-manipulations-in-python-and-opencv-resizing-scaling-rotating-and-cropping/\n",
    "def rotate(image, angle):\n",
    "    if len(image.shape) > 2:\n",
    "        h, w, _ = image.shape\n",
    "    else:\n",
    "        h, w = image.shape\n",
    "    center = (w / 2, h / 2)\n",
    "    M = cv2.getRotationMatrix2D(center, angle, 1)\n",
    "    rotated = cv2.warpAffine(image, M, (w, h))\n",
    "    \n",
    "    return rotated\n",
    "    \n",
    "    \n",
    "# Some of the traffic light images are very skewed. This function attempts to\n",
    "# center the traffic lights for easier feature extraction. \n",
    "# This function filters and image through a sobel image filter extracting magnitude vectors from\n",
    "# each pixel. These vectors are then filtered and an average image rotation angle is computed\n",
    "# and returned.\n",
    "# The magnitude thresold, and angle threshold values were randomly choosen though trail and error.\n",
    "def getTrafficLightAngle(image, mag_thes=(80, 255), ang_thres=(-np.pi/6, np.pi/6)):\n",
    "    # convert to grayscale\n",
    "    gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)\n",
    "    \n",
    "    mag, ang = sobel_filter(gray)\n",
    "        ### print(average(mag, mag_thes[0], mag_thes[1]))\n",
    "    \n",
    "     # threshold between angle values, and magnitude values\n",
    "    ang_threshold = np.zeros_like(ang)\n",
    "    above_ang =   (ang > ang_thres[0]) \\\n",
    "                & (ang <= ang_thres[1]) \\\n",
    "                & (mag > mag_thes[0])   \\\n",
    "                & (mag <= mag_thes[1])\n",
    "    ang_threshold[above_ang] = 1\n",
    "    \n",
    "    # mask the thresheld image with the angle image\n",
    "    masked_ang = np.copy(ang)\n",
    "    masked_ang[ang_threshold == 0] = 0\n",
    "    \n",
    "    h, w = ang_threshold.shape\n",
    "    # If the thresholding only revelas a few pixels disregard\n",
    "    # and set angle to zero\n",
    "    numOfEdges = np.count_nonzero(masked_ang>0)\n",
    "    if numOfEdges > 80:\n",
    "        box, angle = getBoundingBox(ang_threshold)\n",
    "    else:\n",
    "        angle = 0\n",
    "        box = np.array([[0, w],\n",
    "                        [0, 0],\n",
    "                        [h, 0],\n",
    "                        [h, w]], np.int32)\n",
    "    \n",
    "    return box, angle, masked_ang\n",
    "  \n",
    "# CODE TEST\n",
    "# --------------\n",
    "red, _, index = getImage(color='green', Random=True)\n",
    "standard_im = np.copy(red)\n",
    "box, angle, sobel = getTrafficLightAngle(standard_im)\n",
    "# plt.imshow(sobel)\n",
    "\n",
    "# create rectangluar mask from edge detection\n",
    "mask = rectanglemask(sobel, box)\n",
    "mask = cv2.resize(mask, (red.shape[1], red.shape[0])) \n",
    "mask_img = np.copy(red)\n",
    "#mask image\n",
    "mask_img = cv2.bitwise_and(red, mask_img,mask = mask)\n",
    "\n",
    "# rotate image by skew\n",
    "cnt = rotate(mask, angle)\n",
    "# get bounding box\n",
    "x,y,w,h = cv2.boundingRect(cnt)\n",
    "#rotate masked image & crop\n",
    "rotated = rotate(mask_img, angle)\n",
    "cropped = rotated[y:y+h,x:x+w]\n",
    "\n",
    "print(\"angle: {}\".format(angle))\n",
    "f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20,10))\n",
    "ax1.set_title('image')\n",
    "ax1.imshow(red)\n",
    "ax2.set_title('mask')\n",
    "ax2.imshow(mask_img)\n",
    "ax3.set_title('cropped')\n",
    "ax3.imshow(cropped)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c3fc4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIEAAAF3CAYAAAAyxVtiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XmQZed93vfnd5a79DbTs2AGwADgIooxSUmgAzNOqRIppZJNL4ksVxybVZbplBwqC2Wx5Eq8VCpUUlFFf1iWYztRTFk0GceSrESSzaRYjlgqhxQrLimQzFCkyHABAQIDzNYzvd6+yznnzR/dkEbUoO9zMN09PX2+nyoUMD0P3vcs73beu3SklAQAAAAAAIDTLXvQBwAAAAAAAICjxyYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBHyDiPjOiHjpQR8HAOD4RMRfjIhP3/XnFBHf9CCPCQAA4LCxCQQAAAAAANABbALh1IuI4kEfAwAAAICHE88TOE3YBMKpFBHPR8RfjYjPStqJiCcj4hci4mZEfC0i/vJd2WFEfCQi7kTEb0v6Qw/uyAEA9yMi/vOI+IVv+NnfjYi/HRFnIuKnI+KViLgaEf9tRORGmWci4n/en0NeiIj/MiKy/b97ISL+9f3//vP7HyN72/6f/1JE/NOjOE8AwP2LiD8YEf8qIrYi4n+NiH+yPzd8Z0S8tP88cU3SP9zP/8mI+ExErEfE/x0R33pXWY8d8LzxIxHx8/tzyVZEfD4innkApwywCYRT7T2S/oSkc5J+SdL/K+lxSd8l6QMR8Uf3cx+U9Ob9f/6opPce/6ECAA7J/yLp3RFxVvqdV2//rKR/JOmjkipJ3yTpnZL+iKS/ZJT5dyWdkfQmSd8h6S9I+g/3/+6Tkr5z/7//bUnP7Wde/fMn7+tsAABHIiJ62ntG+Ij2nhd+VtL33hW5vP/zpyS9LyL+oKQPS/oBSecl/X1JH4uI/v4LA/+7Xvt5Q5L+PUk/J+mspI9J+ntHdnLAAdgEwmn2d1JKL0p6h6SLKaX/JqU0TSk9J+mnJP25/dx/IOlHU0q39/N/5wEdLwDgPqWUXpH0KUl/Zv9H75Z0S9JLkv6YpA+klHZSSjck/YR+dy64p/13Cv1ZSX89pbSVUnpe0o9L+r79yCf1u5s+/5ak/+6uP3+H2AQCgJPqD0sqtPfMMEsp/aKkX7/r7xtJH0wpTVJKu5L+I0l/P6X0aymlOqX0UUmT/XL+kA5+3pCkT6eUPp5SqrX3wsS3Hf0pAr8fn23Eafbi/r+fkvRYRKzf9Xe5pF/d/+/H7spK0gvHcGwAgKPzUUn/ifYW4H9ee4vtpySVkl6JiFdzmX7v+H8vFyT19Hvnhhe090qvtLfJ8zcj4rL25pZ/IumDEfEG7b176DP3dyoAgCPymKSrKaV018/unhNuppTGd/35KUnvjYgfvOtnvf1yah38vCFJ1+7675GkQUQUKaXqfk4CaItNIJxmrw7oL0r6WkrpLa+Re0XSE5I+v//nJ4/6wAAAR+qfSvrJiHiHpD8p6b+QNNPeK7YXWi64b+3/v09J+u39nz0p6aokpZS+EhEjSX9Z0qdSSlv73x/xPu296tscxgkBAA7dK5Iej4i4ayPoCUlf3f/v9A35F7X36YEf/caCIuLf1MHPG8CJwcfB0AW/Lmlz/4vdhhGRR8Q7IuLVL4D+eUl/PSJWI+KKpB987aIAACfd/iu3/5ukn5H06ymlr+9/TOyXJf14RKxERBYRb46I75hTVq29eeJHI2I5Ip6S9MPa++6hV31S0vv1ux/9+r++4c8AgJPnX2rvHTzvj4giIr5H0rsOyP+UpP84Iv6N2LMYEX8iIpY1/3kDODHYBMKpt7+A/3clPS3pa9p7VfcfaO9t+pL0X2vvrf1f094Dwj96AIcJADhcH5X0Lfq9Y/pf0N5b939b0h3tbRQ9apT1g5J2tPelz5/W3ubSh+/6+09KWtbedxHd688AgBMmpTSV9Kclfb+kde19fPj/0N67Ru+Vf1Z73wv097Q3h3xF0l/c/7t5zxvAiRG/9yOQAAAAD7+IeFLSFyVdTiltPujjAQCcfBHxa5L+p5TSP3zQxwIcFd4JBAAATpX9X9X7w5J+jg0gAMBriYjviIjL+x8He6+kb5X0zx/0cQFHiS+GBgAAp0ZELEq6rr2P+b77AR8OAOBke6v2vvdtSXtfCP3v73+HHHBq8XEwAAAAAACADuDjYAAAAAAAAB3AJhAAAAAAAEAHHOt3Ai0uLqZz51bn5pqmscts6trKzWaVlesPBlZuOBxauTbZLPP25CLCrttOHvLHAtt8zNA+RjPoVt0kv51NJ1MrNxmPvfKmXnmSlBrzhMwTb2TmWvTDzGyTZVlauV7Zs+suCm8Yy/Lcyrltt02Pccq8fvOGNjY3/c59Sp0/fz5dufLE3FyLYfDQuWNMajHG1LWXrc05r669Oa+Nw56j2sxlh/3R9TAnlCzzj7E2x8zJ+J6/efge5Xn3WvLvjTsG9/t9K1fkbZaRhzu2thkCGnd+NO9hXXn9azabWTnJaz+3bt3S1tZW5+eJM6vn0qXHrhjJI1gvPwTc8a3NSbvjtduHGnPOG0+8dbXkr9Xd8bIwx8vh0Ht2lFqMmfYzjz83JvPe2HPzA+w0btVH8aU39jU3c23WQs7i95WrL2r99trc4H1tAkXEuyX995JySf8gpfRjB+XPnVvVB37o/XPLHY/8Dr+14f3SjxvXb1q5N3/zW6zcO77lW62cJL3jW7/Fyi0uLlq50nzolaRMXofPanNxZj6AuJtzUouHOTNYJa/u0e6uWbH09edfsHJf+9JXrNyLZnmSVO16k5q7KN2tvP61OxpZOUka9r0J8PLlR63cE487i7s9Fy5ctHKLi8tWbmZOklWLNl4ZZf6nf/WH7fIeJm3niStXntAv//InjHL9N7K2eZB3uBsxY3NTWJK2tras3OamN+dtbNyx63YX7e4c1et5m7hHsQnkzj3uQrzNCz7uPfzqV796qOVJ0sLCgpW7dOmSlXvLW7y10LnVc1ZO8l94ce91m/bjvvCyvb1t5e7cuW3lrl27ZuX26t6Zm/mvfuSDdnkPk7bzxKXHruh//JmPOyXbx+BuDCiZG8h2zb7D3mDPzRfHJH9DZGyurbe2vLnsS1/6spWTpBde8NbWCwveXHbp0iNW7h3veJuVk6TV1flvhpD8e1i12Gh2x0G3XeS528r9F8Ts/Scz2GqTzMxW5vOWu1Z0X8yWvHvz3j/93VZZr3uMiohc0v8g6Y9Jepuk90SE3wsAAKca8wQA4CDMEwBw/O5no/pdkr6SUnoupTSV9HOSvudwDgsAcAowTwAADsI8AQDH7H42gR6X9OJdf35p/2cAAEjMEwCAgzFPAMAxu59NoHt9GO/3fZguIt4XEc9GxLM7xuedAQCnRut54vbttWM4LADACdF6ntgwv5MJAHBv97MJ9JKku3+FyxVJL39jKKX0oZTSMymlZxaXvC/iAgCcCq3niXPnzh/bwQEAHrjW88SZFl9IDgD4/e5nE+j/kfSWiHhjRPQk/TlJHzucwwIAnALMEwCAgzBPAMAxe92/Ij6lVEXE+yX9n9r7lY4fTil9/tCODADwUGOeAAAchHkCAI7f694EkqSU0sclfdzNN02jndF4fm5W+weR5VasNxxaufFkauVu3Lxp5STp+RdesHLnL3hvb11c9D9WVxbe9emXXlNwy8tK/01mVeXd7+nurpUbbW1ZuZvXb1g5Sbr6lees3I3nvXs9uuG3n8XewMqdGXi5bNCzcvXispWTpKH5Uc9HLl22chcfuWjXvXTmjJXL+955x2xm5VLlt/FUN/Prze71tQgPv7bzRESoKOaPRxH+9crMazuZTKzc5uaGldtu8T1447E3vu3sbFu5LXMclKSyLK3cxYtevzx3zpzLFhasnCRtbXnn/dxXv2rlNja8e+i0xVf57WfTyrlzoyTVtZdtmvljkSQlM3ePr255TVnmjZkpeWW65yL5bdxtuysr3rzzyCPenCdJVTV/7mmz/nuYtJ0n8izXorFGcfuFJM3Mud/OTSsr5/YLScrNeS/MMovC778peXXXxnpnrzzvGOvaP8apec2lkZW6c+eOlXvhhRfnh16t2XgOlqQzKytWrmeubSVpYD4nqMX6yipO/ljtj/9u3/bPJcJrk2XpPQuXpXcu5ikfuvv5OBgAAAAAAAAeEmwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdEBxrLUlSc382NbOyC5yY2Pdyu3u7lq5pfGKldve2bZyknTzxg0rtzvasXKDYd+uuyhzK7ewMLByvZ7XZLxa9zTTqZWbbmxZua3rt6zcK1993spJ0ivPvWDl1l/27nU+q+26z5y/YOUunvfu4XJ/wcqVQy8nScMzy1ZucfmMlesVpV13qisrNx17uaYxBilJKSUrJ0lJfrbrUmpUVfPHhIiwy8wy7/WOpvH6pXvvs+zwj/Gwc5IU8o6zqb3rU8+8vjabzqycJMm85v2+Nz+69/DatWtWTpLW1731yPbmppXL8xYzaXPeis1WV73izHvtt3ApM/vsUYyWjTtem8fYK721UL/nzXl7dc+PFMXxLttPrpDS/DFud+Q/T9y8edPK3blzx8ptbGxYuTZjdRZmtk3HdJl9yB233LZcm2ORJF26dMnK9XvePFGY/XzDfD6RpM1NL+sucYoW6+Wy9LK9fs/Lmfcwwh/V3fMOc33lrgkkacF87llY9HLDwdCruMX18SZI79rwTiAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqATSAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqATSAAAAAAAIAOKI6zsiYlTaezubnd8dguc3vkZSeTqZVz694djaycJG1vbFi5anfXypWlf9uK0tvnu1nPvy+S1Ey969i0uIex4513veZdx8n1NSu3ffW6lZOkdOO2lTu75Z33cjmw6z6z6eWWbnp1Ly4uWrnh2TNexZLKM2Z/uLllxXaGPbvu3Z7Xxkd5snJVv7Ry0zysnCRNjehsMrHLO80iQnmeH2qZWea1kV7Pa3erq6tWLs/911k2N73x7c5tb3y7efOGXfds5o3/a7e8MgfDoZXrlV5fa6Oe1VZuZ2fHyo22t+26x+a6oJpVVq5pGrvukXk+63fuWLmbN7x7nYXfxhcWF6xcUXhrnLr27nUb7liRF94YVRR+G3eupT/rnG51XWtzc/56Yn193S7z2jVvTXj7trce3Nz0Fm8L5ngpSb1+38olc+jwVkV7ssxtfYfbStvMo2XPuz5Z7o0xYb5Xoq78sdod193xrar8OWpWeXN9Sm7L8HLRokm4dSezkffMNiFJi+YcdeXKFSv32GOPWrm+2a8lb04Jsw/yTiAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqATSAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqATSAAAAAAAIAOKI6zsqqqdPPm2tzcrK7sMsuytHJhlldPZ1Zu586GWaK01eRWbtbzcoV7MpJi5l3L3a1NK1dtbVm5ZnPbyklSuuOVmZm5cmNk5ZYnjZWTpMfNe7gcS1ZuserZdVe3xlauubZj5VLu7f2Oh30rJ0mjgdcPJ2Ybnwz96zPqe+ez3fM6zqTnDYvTvn+MtVHmbLRrl3fapZTmZiJaDIRGeZJUZOYY3PP6xm6Lezoeef183Zx7bq+t23Xv7nrH2e9580SePbjXl9x2EfaqwGs7kjTsD6xcWXjjZV35a6HRjjfvvfD8C1ZuzVirSdLCcMHKSVLPnFNKcwyO8NtZ5vbtwssNh0Mrt7KyYuUkaWC0n91db5zogjDWMqnFNDGraytX1d7a0Ywpy/21RNnzxpjZzHuWaXF5VBSH+8jozPOSlBp/DJ7NvHs4q/z1v8U8F0nKzIuemXOZO7ZJUq/0xkz33tjzY4uGVpt7AOOJNxbOdrxnI0naMbPLy964fv78eSvn7mVIktUs3DZm1woAAAAAAICHFptAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0QHGclaWm0Xh3NDeXF/5h9YvSy2W5lasmUyt369oNKydJW9dvWbm8qrzczMtJUj6ZWblsd2zlyrF3ffpjr15JKjbntwlJ6k9qK3cmeff6Yn/RyknS+aUlr+6hV+ag6Nl1K6XDzZm2d3ft7NbEaz/THe9ex7ZXniQ1yWsX02pi5TbllZetrlg5SRpcODc3E7PGLu+0y2L+6xMRYZeXZ97rHZmZk7y6m9prS5I0NsfgkdmHJmZ5ktRU3nGWQ29u7vf7XsUthqxkjm+p8fpRY5bXmOVJUpjtojDbbp77r9OVmbcWcq/5ZOS1n8muN65Kkszzicy7Ps448ara7IuTiXc+hblOXVnx5wmn3+yOvP5/2pVlqcuXLs3NtWkj63c2rdxox1sbbW5uWbmyZ/ZdSYPBwMq57X0289fqqTEHD39qNotrUaAZdUt06/bXDn7lyZwnIrWZow6X+9iRtViv1fb66nCfeSSpKLznR3c9Mp16/auuH8z6n3cCAQAAAAAAdACbQAAAAAAAAB1wXx8Hi4jnJW1JqiVVKaVnDuOgAACnA/MEAOAgzBMAcLwO4zuB/p2UkvelNwCALmKeAAAchHkCAI4JHwcDAAAAAADogPvdBEqSfjkifiMi3nevQES8LyKejYhnx+MWv0UCAHAatJon1tbWjvnwAAAPWKt54s7azWM+PAA4Xe7342DfnlJ6OSIekfSJiPhiSulTdwdSSh+S9CFJunD+/OH/PjcAwEnWap54+umnmScAoFtazRNv/7ZnmCcA4D7c1zuBUkov7//7hqRfkvSuwzgoAMDpwDwBADgI8wQAHK/XvQkUEYsRsfzqf0v6I5I+d1gHBgB4uDFPAAAOwjwBAMfvfj4OdknSL0XEq+X8TErpnx/KUQEATgPmCQDAQZgnAOCYve5NoJTSc5K+rd3/FSqymJvqZbldYlGYp9B4Hx/e2dy0ctubW169kmY7Iys3MI9xJSvtui/0FqzcmeSVWc5qK5c2vXOWpHzUWLnVvncuF1bOWLmVpUUrJ0mLq16Zw9WzXm55ya47cu8Ne5F5uSzm90FJatbWrZwkNZvbVq40vxx+MvG/RL66ddvK3bx9xytvNrVyq4tem5CkRxfmt4s2497D4vXNE1IyhsKQ1473uG96dXNu3f6bbbPw7n9mlpm8YVWS1FReuHZzuZdzxyJJe18ba4jklZmZ5bnXW5KSWag51Uvm2C9J0eZaGpLTCWXflr2ssf6TpMw8b3v9J2k2m1m58e7Yyrlz1M7OjpWTpOl0/txTNy069kPidc0TIeWF0Z7cji6pSpWVq2Xeg9xr773hwCtP0tBct44mXjtuKq9fSJI5tModig53xDqqMg9/HHSP0R2Dj8Jh15xalJjn3lpoacl7jpqZa/o22dHIe8bd2vL2Cs6e9Z8nhsPh3EwW5vOgXSsAAAAAAAAeWmwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAcVxVpZSo8l4MjfXW/APq1BYud3RjpVL08rK9cLfP+v1+l6uaqzccuGVJ0mPLJ6xcuejtHJptmHltmZbVk6SZtPayvXOLVi5C9/0lJV75E1PWjlJWn3D41ZuePmClYuFnl33RN71qc0mWZbevZ5eX/MKlNTb8PpXf5Ks3M3nXrLrvvXZ37ZyOy97ZUbujT9LC0tWTpIunD0/N1MUxzocn2ChzBhfI7yxv1U2ee2znnnzRD2defVKSo03/rtn7V8dSfLOu6q888kzr/Yyz62cJIU556bGOxc357aJNtHwi7S59zu5B2mft38y0bJVzq/ZmxslqZ5Ordxkd9fKRea1x94Zb76VpOHC/DVOZtbbBY3TnHK/zRWlNx65t6BpvPZZ1f48UTXmGGyeS5n8dUdtjpnu+OaPBy3GYDMb9iDsHaNbbxvu9bHHdPlX0r+OXnnmkmC/THMdVntrprr21muSNJmMrdzu7sjKjXa3rVyEP5f1B/P7trtFwWwCAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAcVxVlZVlW6u3Zqb29rYtMssFFZutLFl5VLdWLle+PtnS72elVss+1YummTXvX77tpUbb0+8uje2rVza2rFykrQ4GFi581cetXJPvvPtVu7c27/ZyklSceWClUurC1au8pqEJCl6ZlvLvb6wI6/9lG+66NUraXFUWbmFndrKjYb+0JTdumblpl/2ypyOZ1ZuIm+skKQ6m39v/F592iU1zfxrm2f+GBxe11BmBpN778NvI00ztXJ17eUkr6/t126lovHKjGSWF7mVkySjC0mSUub1pOT2uOT3TDdpnko77nGauWTewzbjVjK7bJj3MFr0L7eNp+TNZUVWWrmFgT/ZryzNXz+0GfdOs5Sk2lmvt2igmTvImLegNsfg0XjkFSgp3/Yqr2qvbnsclKRklpnM62jOt+78LUlhjq6HPaq3GdPtulvMPYfNvebueUerm+i1cXf14M5lkjSZjK2cO1YsDIdexS0uz3Q6fw2YzLbDbAIAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0QHGcldVNo+3RaG6u6dV2mQtFz8r1+n0rt72+YeV2duafx6s2m8bKZefOW7l+f9GuO596dY+2vfOOnR0rV1aVlZOkpeGyV+bCwMoVPbNZt9kCrWdebjaxYtEv7aojMw+0CC9nFpcnr2/tFekV2lTe9Vm57PUFSXrkDY9ZuQvPP2rlrl29buWagT98xoJxLTPz/p12SUpp/hyQlOwi3faZzLF6PB5buZ3tTSsnSVubXna86809WXjnIkn9Mrdyvb6XK0vveudecZKkzLzfKZk5s16zuFalRrtCPebw4VZtH2KbYcsd48w5Kslv40nemsTNKcw23mKVXZTzr08wTUja6+ez2XRublaZazdJtTHvSFJj5px5TJKaxl8vV7XZju0O7I9F7pxrdo12Y8cDYl+dpsWYfgTDvysOewCxi/PH6tp8fpxMvOcJZ5x4lXt9FhaHVm5pxXteX19ft3KStLExP+teG94JBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AHFcVbW6/X05BNX5uYunb9ol3nx3AUrl0dYuZsvv2Llrr34kpWTpBtXX7Zyt9bXrdx28nKStBK5lRtMKyu3UCYrt1SUVk6SNmcjK/f8i1+zctH36n174bUJSVqpHrNyxeSMl9vy7osk5QveCfUGPSs3LL2938j8e5hV5rWc1Fas2dq2656a2dlkbOVS7rVx+ZdHycn6zfF0Cykzuke0eQkjvHs6GXtt5MaN61bu5VeuWjlJWrt908qNdnesXJ75F6jse415OBh4defe+NY03niwl22sXDJzktcmwlw7tCkzNWYumWNRG2aZ7nUM815LUp6bS07zms9mM7vupvbbmiMz+5fbFyQpL+ZnWzXH0yw1qmaTubGm9ttIZl7blLy2NJtNvdzUy0lS1TfPx20nbYYY+/ocwbhlOuya/cvo12z3YbfII5ij7HnPrdp8FpWk3Fy6FIV3Llnm968mefNeXXnPzHXljRWDxQUrJ0n9/vxnPXfe4Z1AAAAAAAAAHTB3EygiPhwRNyLic3f97FxEfCIivrz/79WjPUwAwEnFPAEAOAjzBACcHM47gT4i6d3f8LO/JulXUkpvkfQr+38GAHTTR8Q8AQB4bR8R8wQAnAhzN4FSSp+SdPsbfvw9kj66/98flfSnDvm4AAAPCeYJAMBBmCcA4OR4vd8JdCml9Iok7f/7kdcKRsT7IuLZiHh2Mpn/JW4AgFPhdc0Ta2trx3aAAIAH6nXNE3duM08AwP048i+GTil9KKX0TErpmX7f/LVNAIDOuHueOH/+/IM+HADACXP3PLF6jnkCAO7H690Euh4Rj0rS/r9vHN4hAQBOAeYJAMBBmCcA4AF4vZtAH5P03v3/fq+kf3Y4hwMAOCWYJwAAB2GeAIAHwPkV8T8r6V9KemtEvBQR3y/pxyR9d0R8WdJ37/8ZANBBzBMAgIMwTwDAyVHMC6SU3vMaf/VdbStbGC7o6W9559zc2TNn7DKXFhas3Hg0snKrK17db3zqDVZOksZbW1bu+ksvWbnb167ZdW/f8r48b1yPrdyomnnlNVZsT/LCvXXvXPIvVVauGPtfVH7lpaes3MUnHrdy/Yv+59ljZdHLDXtWLsvDqzgrvZyk2YbXxneee8HK/dZvfsau+4tf+P+s3PUXvf5VLHjXO8utmCSpaqZzM0nJL/CEOcx5QpIi5l/cLPw3smbmzYrMLNPsQqnxB8Kq8satpvFyvb43HkhS2fNOKM+988nCa8sh//ok1V6ZmVl3Mvtbm27plmkX2qLyZDZKk9u93OstSVnm3e/K7DfT6a5dd5p4a5yeec2Xh953XF48t2rlJOnCxQtzM2Xpz8snzaHOEyFlMb/NF7k/TxSFN08UuZfLM69PphbjYF17438Y12Yv12IeNcs87JVMu/KOYFw/bId+gfwC3aQ7m4Q576TktVvJX+Nh7iSRAAAgAElEQVTUtbcmyAt/zFxeXrFyly5dtnJPPvmEXbdv/l3MzLXskX8xNAAAAAAAAB48NoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOiA4jgr65U9PXnlibm5ajazyxyPJ1ZuOq2s3OrqqpVbemxg5SQpzby6H7l00crtrN+x697d2rRyv/Grv2rlbl+9auXGVbJyklSo8YKTqRWbbmxYudHNNa9eSS996StWbmV5ycqVg75d93DFK7Nnlpk13vWezWorJ0lbW9tW7uaad83vbG7Zda+PdqxcmONKkYWVy3IvJ0m1/P7QdRFSZrw8kRf+axhlv7RyVe3lcreNtHiZpci9XFl4dffMnCSV5nHmyRsT3NMOd+yXpPD6UJh9zc75l1GZDnnsaFqMMZV3b5I5FPk1t7hA5jVvGnPuqbw1gSTlyVuHDc2OuDzw1oBnlxatnCStLq/MzeSZOVCcchGZ+v35a56i9MZ0Scpzb+TKzIE9z7175ebaZj1t+q+XdcfM5A5GLZZPbc7mMEVz+Gu8xrw+R3POZqlmLDPXTHtFell7nkj+OsNdsw2GPSu3ZI7/k8nYq1jSzNxTcPBOIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA5gEwgAAAAAAKADiuOsbDqb6utXX5qbK8vcLnPY71u5pZVFK5ciWbm19TtWTpLu3FqzctVkbOUGpX/bLj9xxcpdectbrNy0qq3c2tWrVk6Sqto77828tHJnMu/6bM62rZwkLax52cENr/2UM+86SlIv8/pDL7xc3nh7v3VYMUnSRN5575q5SeFXvpUqK7eZeXUvLfas3GjotUdJ2urNP5821/s0C0m5mrm5LPl9KFVTL2jmSrMtlS1eZinM88kbr73nLab30uyX7nnnZltOxn3+HY1XdyQzZ1bb5pWyzCw0C6/UaDEmJPNIG/M6JvM6phbHGJl5jOacl9l3USpyrz/0+15ueXnZyq2YOTeb57x2K0mRhcrB/Dm47PnzdOa2O7Mdh9mB3fLaZJvGG1vdfi61GI/MQcF83DoSbt32Kbep3K37AV4f/8S9g2xqf73mZpvGy1XVzK67qr3s9rb3TLi5uWnlyhbP9T1jTLPHHrtWAAAAAAAAPLTYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAOK46xsMh7ry1/84tzcpUuP2GU+/thjVq5XllZue3vLyu1sb1s5SUp5WLnVRy54uTMrdt1nVpat3MW1m1bu5ds3rNz2jatWTpKm9czLeZdRs8ILblZevZJUTmsvVzdWbiDzZCQNwmu7/dzLuX2h8Q9RdeaFqzy3crvyrqMkjZK3l71r7nlXUVm5tHXHyknS1kvzz3s8ndrlnXZZzG9PkZJfYO3139lkbOU21tet3PbGhpWTpNR4xzjo9azc8uKiXffK8pKVO2PmhoOBlSty/3WoXuktV9zxrci8usNoi7/DbZKNF0xmTpKaxhsz68ob36bmeDTanVg5SVrf2bVya2b/ctcOkj++9sz7ncxmkRVee5Sk0ujbEbx2K+31DaeNTqd++3TbvJubzfz26crNNVSy58cW82iLdatVWpux1WWet1t1mJfnSM7FLLNNzYd9lO55hznf7pVp3kNvyaS68eY8SZpOvTXg5qa3tltbu2Xlzq2etXKStLAwnJtx7zOzCQAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAcUx1nZbDbTzRs35uamk4ld5mQ89nJmmcvLy1bu0qOPWjlJWl09a+UWFxesXFnmdt2zamblmmFp5XayZOVuN1MrJ0nRNFZuknl7lvXAa9bDFlugMam83NS73qV3GSVJg9I70EHPO+9er2/lWhyiajM9a8xSS+8YJSmV3nmH2W92FFZud2fTyknSzdn88Wcy8/vMqZakZAwJKfc7cIQ71XltpKq8dpwVPbNe6fxFb04ZDLwyn7jyuF33uXOrZt1ev3TnqMzrapKkPPPKLAuz7tyfR10pee0imXNecsfLNnWbucY8xrrFMc7MMtdu37FyX3vua3bd1421pyTt7OxauWS28d3aX89ujnfmZupU2+WdZhFSYfR1J/O7WW+eyM2xI8Ib4MrCW39LUq/njf91bfbz2m9P5umYKyhfm/LCTdsxL9jqnN26zbG6VdXugZpBt7w898+lDm9tl828yjP7pH1hPvP418dfz1rjlFkx7wQCAAAAAADogLmbQBHx4Yi4ERGfu+tnPxIRVyPiM/v//PGjPUwAwEnFPAEAOAjzBACcHM47gT4i6d33+PlPpJSe3v/n44d7WACAh8hHxDwBAHhtHxHzBACcCHM3gVJKn5J0+xiOBQDwEGKeAAAchHkCAE6O+/lOoPdHxGf3397pfaskAKBLmCcAAAdhngCAY/Z6N4F+UtKbJT0t6RVJP/5awYh4X0Q8GxHPTvntNwDQFa9rnri1tnZcxwcAeLBe1zxxZ+3WcR0fAJxKr2sTKKV0PaVUp5QaST8l6V0HZD+UUnompfRMr/R/XS4A4OH1eueJC+fPH99BAgAemNc7T6yev3B8BwkAp9Dr2gSKiEfv+uP3Svrca2UBAN3DPAEAOAjzBAA8GMW8QET8rKTvlHQhIl6S9EFJ3xkRT0tKkp6X9ANHeIwAgBOMeQIAcBDmCQA4OeZuAqWU3nOPH//066ksslDem1ulVs6u2GU+9thjVu6Nb3yjlVtaWrJy/X7fyklSWeZWruh5H5fLC/8NXHlTWrnllbNWrr/oXZ8qvHOWpJRqKzfL57cdSdLighUrWtzDvElWLmaVl6sbu+5KXt3bbnnh1V20+Pim23az0ruH08q7jpLURFi5yLy6s9LrM03mt/E65vfZJO88TqLDnCeSpMZ4k2okfxxs5N2rSe31ta2dXa/eFm3kiTe9ycq5H5e7cNH/WN1g4I2FVTWzcnXt9d+68cZ+yR8Hp15Mqs26syPol+aY1bR4r3btzinm+ZS9gZXr9/15Ynnozc2LZtvtr3jrEUk6d+OGlbtx46ZdpiN6/hgwqed/b2ZKbgM/eQ77eaLfnz9XD4dDu8zl5WUr1+97fSPMeWdo9gtJWjSfUXZ3x1ZuZo7pklQW5hrcHN/MId2N7VXt5g69G/nzhJ3MvAkgWlwhP+veHC/XmGsrScrMK7Qw8Pp2U/nrjNGO9yQ1nXjfcTw2+2FjPmNKUp7PH1fcLng/vx0MAAAAAAAADwk2gQAAAAAAADqATSAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqATSAAAAAAAIAOYBMIAAAAAACgA9gEAgAAAAAA6AA2gQAAAAAAADqgOM7KBoOBvvkP/Gtzc49evmyXeeXxx63c2XPnrFwe3r5YKFk5Sarrxspls8osMOy6q3pm5abjqZWb7Zq5sVevJPWL0sotDJas3PLSWSs3KPzmnyfvfkfj5TK/+aipzHtotp+J2SaSeV8kKXo9K5eZuZ3dDbtu97xLs+7l/tDKJXOskKSqmn+MqcWYcrrF/j8Ha3O13GtrdnN5I7qUtRhjFpdXrNzS2TNWrj/02rEkZbnXlt2xqHYrNuuVnBaxx24XZoER/nzrco+xTc1F6bW1cNc4Zq5p8VqiMw5KUpgXaHFx0a77/PkLVi43+2xTewe5uOitWySpV86fo9z7gj1t+m9ujkd5nh9qrk1HT+Yas7Fz7mx2NOOWVd4RjMFhHqXd21oco//86N8bl7vGOWxt7mCWeVfdnfMm04ldt9vWJhOvzO3tbSvXNPaqSUUxf1xxz4PZBAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAOK46ys7PX0+BNX5uYWFxbtMiPPvVzm7Xcls96mbsyklMxsZu7JTaqpXfed27et3EvPv2jlbl27ZeXqaW3lJKnsL1i5Xtm3ckXes3J5UVo5ScojrFzmNqDkBiXl3nFmeWXlimpm5aa1fw+3bm9aue3RyMoVpX9vLl2+bOWeeuoNVi4vvGHxzvq6lZOktbW1+aEWTeJUi6QU88fM5HXJ/TK9MbiuJ1Zue9tr7zs721ZOklaWvXFwccEb3xYHfh8qB97YWpovG5mXW03jz6My73dmzvVhjumtxmqTXbd70pKa2jvOpvHmicoc/+vwX0tMfa+dJbddtFhnZJVXZpm8NaVy77yHhXfOktTrzc9mWZuB7/RKTdJsPL8tz6Zee5ekqjLbkzkk5ObzSV35xzidenNUbrbPXs+bT6TDH1v9lnz4SbvEQz8XKey02dBazFHJLDO5ZZq5NtenNp+Zq2ps5WYT/5nZ5rYLO+fPo22y8/BOIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA4ojrOyLMs0XFiYm8sL/7CaprFydVNbuTDrrSuvPEmqZzMrV+TeeY/HU7vum7fWrNwNM7c12rFyUZZWTpLKwdAM9qzYNLy7mGe5V6+k2mwZSV57TMnL7dWdrFyTmec99K73xprXJiTp1tptK3f7zrqVe+tb32rXfeXJN3hlvv1tVm4y8frXQovrMzDGvfKz/8ou7zQLSYXRNbMWL2G4Wbf/1vXEym1t3fEqlvTS1712V01GVm5Q+heol61auTx5Y1FhjkXRYq4Pr2ol8xhl5tw1gSRl5tyThTf3pMY8F0mzqvLKNM87N9cjeeHP9VnhzeGbW9tWbu2Va3bdt27esnLbO17/Gg4XrdzyU/PH/t8pc2H+tcxatchTLEl1PX8dPjPnc0kajbx779QrSYU5vrlrDkmKzFyDhzf+93pen5T85yhzySqZ42WbFh9mOsy6/Xr9sdov033u8Lnjvz2PmrW77XG/civm9sPU4gq5fbbf71u5gft828JsOn9PwV078E4gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAOKY60sz3Xu7Orc3HQ6tcscj8dW7tq1a1aumlVWrqlrKydJWebttZ1fPWfl6so7Rkkaz2ZWbms0Msvz6s7y0spJUpQ9K9fkuZWbqLFydeW3s6YxyzTbRUrJrtuVKaxcbt7D0dRrO5I0rs1+k3vH2F9etOsenlm2cr1Fr8y68IbFvtm3JGnRuN+ZWe+pF1LE/HYSZnuXpEhm3zBfFykyc3zzpwltbWxZuWrsjVtbGxt23WdWvD60vLRk5VYOOSdJg743T7jjoMwx2B3798JezG279rlIivDarrt22ZlMvJy5dpCkOxubVu7GrVtW7urVl+26R+ZacWG4YOUuP/qYlXPHHkkq8/788sz7fNrVTa3NzfntaWe0Y5fpPntU5hrcXee568a9ur1slrlrTL99psYr05m/22i1Wj7cqu3iWnRzu0x3nmjzPOHemxRemcm9O62ujzfGZZn3TOjmXq3d0e8NrNzSorfGKQtvfSNJWRjnY97nuVc6Ip6IiH8REV+IiM9HxA/t//xcRHwiIr68/+/5uzsAgFOHeQIAcBDmCQA4OZzttkrSX0kp/QFJf1jSfxYRb5P01yT9SkrpLZJ+Zf/PAIDuYZ4AAByEeQIAToi5m0AppVdSSr+5/99bkr4g6XFJ3yPpo/uxj0r6U0d1kACAk4t5AgBwEOYJADg5Wn24OCLeIOmdkn5N0qWU0ivS3sAu6ZHDPjgAwMOFeQIAcBDmCQB4sOxNoIhYkvQLkj6QUvK+3W/v/3tfRDwbEc9uGF/iBgB4OB3GPHHr1trRHSAA4IE6jHli/TbzBADcD2sTKCJK7Q3Y/zil9Iv7P74eEY/u//2jkm7c6/9NKX0opfRMSumZMysrh3HMAIAT5rDmiQsXzh/PAQMAjtVhzRNnzzFPAMD9cH47WEj6aUlfSCn9rbv+6mOS3rv/3++V9M8O//AAACcd8wQA4CDMEwBwchRG5tslfZ+k34qIz+z/7G9I+jFJPx8R3y/p65L+zNEcIgDghGOeAAAchHkCAE6IuZtAKaVPS4rX+OvvOtzDAQA8bJgnAAAHYZ4AgJPDeSfQoYkIlcX8KmeTiV3m1MzOZrNDzaWUrJwklUXp5cotKzcej+26X7l+zcrdWvO+ZG/XrLs36Fs5Seq72fC+x3wy9u7hRF5OkpqmOdTca62C7pnNvPPOwyu1MCufVP71mdW1FzTPpWzRfrJ+z8pVZnlj87xnyTxnSSk3LnqbRnGKhTKV2WBurmn86z8eT63czo43vo3dMWbitjop5M0ps+mulRvt+vPojeve+J+ZY8zA7JMry0tWTpKGg/ltQpIKc4wJcwqv3bFNkhp/XeBItTefSFI9847TXgtNvdzULE+SJpXXD7dHO1bu9p11u+5+32s/Fx+5ZOXOP3LRyq2snrVykrR0ZnluJstb/VLfU60x5tXcHIskaWll/vWXpN6dO1auNgeZSn4/z81sZo6DbbjDm1tzZpYXLdp8Mp8T3HvTmOsyN9eGe32KFovH3F2qN7mXM5+FZ5l/jE3mtfHKnPOaFn1hOFzwcj3vGaWfedexmfjz6G7Mnx/dtQOzCQAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAewCQQAAAAAANABbAIBAAAAAAB0AJtAAAAAAAAAHcAmEAAAAAAAQAcUx1lZSlJT13NzEX6ZRVFauTz39rvKXs/KRYuD7JtlZllu5ba2t+26X375ZSu3vbVp5dyzXlxYNJNSv9e3cu6OZT2dWrmUklmi5J65e33atJ/MPU7zAiUzVxl99VWzqrJyjbxzyQuvL0hSZN61rOqZlZtMx1ZuWnntTJLqZFzLVu3x9IrIVPYGc3NTs59L0sy8p5Ox10bq2rtXlZmTpJnZntyRo1d6c6MklbnX3zJz7BhPvHNZu7PuFShJdWPF7DHYvDUtliM+t6+3GBJS44XduSdzb3aLuSyZJzSZefNJMsd+SSoH88cUSVo+e9bLra5auf6SvxbKB8O5mXDvyymXZ7nOLSzPzZX+Mkbj2LFypbmIys26c3dRJilzF3recGmvB/eyXrgxx4Rk56zYftYbY2rzvGuzbvecJf983BKzzJ8o3Nsd5hwVjdfQ8vAbmjs3N+YVSm1mcfPmLC4uWbnz58571Sazw8rbR3HnWmYTAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADmATCAAAAAAAoAPYBAIAAAAAAOgANoEAAAAAAAA6gE0gAAAAAACADiiOt7qkpqrnpkJhl1iW3ilked/K5YVXXmHm2mS3tjas3PrWul33K9dftnJVmn9fJKkc9Lxcv7RykpTUWLm6qaxcJK/eLPx2puQVmpJXZpuqs8ILZ3luFujFUu4fZJRe3aW8nH0ukuraa7u7u7tWbmdr28vt7Fg5SZpOJnMzTeP1g9MvKTWzuammntolzmbzr78kTadjrzzjfkrSbOyVt1e3dz5hDh65ORZJUmZOZykzxyIzV5jz8l6h5hjceOOBO6YfhRbDvy8d7vhhl5a1eC0xvIaWzEmqzuaPE68a1979vr3pjf8rd7z12uq5kZWTpH45vz8k5glJe+Ngr5y/Hi1yf60e7pjgDh3mWN1mvZOZzxONOcok+2Skxhxj3DkqMnfN6o+Y7hzVmEU25vDm5iTJnZrd5phl/piQh5fNcu+EovEOcipzXpbUmGt699mxxaOMysJrk9Paq3s89uaovrmXIUmlMV64+yi8EwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADoADaBAAAAAAAAOoBNIAAAAAAAgA5gEwgAAAAAAKAD2AQCAAAAAADogOI4K6vrWltbW3NzEWGXmRfeKRS93C7TsTvetbPbO/PPWZKuXb9m5Z7/+gt23Xe2Nq3cTI1XYOntGzYtLve4nlm5vPHaRR7eMeaZvwea3KDZdN1jlKRwjzP3Kq8z72xS2aIfDrx+mOVew0iFX/csee1nd+r12fHEy1XTqZWTpNSY/Qu2lOxeqca8/nVdH2quqvz7Pp157djtGbnZ19rIzLHIzeXmWCS1WBeYY6tdnH3F/XsTYY7BLdp4MpuaW6a/DvOvj3s2bn+dmX1G8vuDW7d7HdvcQ/hSSJNs/r0ahz8GT8x+OTHLdHN5ePOJJCVzrV6Z3bJp8byl3FyLZl5fC3PN2uYYm8xdB5t1m8eowp9vk1m3zLGjqVq0n6yycuGuXdw1U5t5wp3Dk5fr2TVLTfKO89aG92z99atXrdzlixetnCSdWVqaHzKnHd4JBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AHFcVa2u7urz332t+bmirK0y+wP+lau1/dyu5OxlVvf3LByknT9xjWvzPV1M+fXvbHhZeuqtnJZhFdeSlZOkho1Zs7bs2zCK6/2D1HeWUsR3jFmub//2mRe7WEepXuMRdGzcpLU63vtpyy8ISe1aD/j8cTK7WzvWLnpdGrlmsY7Z0mKaNHYoGS00WT3SqluzDHBznn3s05eeW3KdPt5iy4kN9qYhSbzOjbJv4du0h873OvdRrv0fP5NtE/bPERz2lGYawJJiszsN2b7mc4qu+689MbryHKvPHN+zFrMo1nurH157VaSlJLX19u0z8K79yn3yqzMeX+U/LXERDOvbnONGT3/easYDK1cbpbp1t20WC/X5sA1M8eimXmvG7PtSFIyj9Fd47SZdczHI5VmqZm5btFo18tJasxss+Pl6h3vuV6SZo2XnfW9Z5lsccHK9Ze9nCQtrMzPus+YzCYAAAAAAAAdMHcTKCKeiIh/ERFfiIjPR8QP7f/8RyLiakR8Zv+fP370hwsAOGmYJwAAB2GeAICTw3k/UyXpr6SUfjMiliX9RkR8Yv/vfiKl9DeP7vAAAA8B5gkAwEGYJwDghJi7CZRSekXSK/v/vRURX5D0+FEfGADg4cA8AQA4CPMEAJwcrb4TKCLeIOmdkn5t/0fvj4jPRsSHI2L1Nf6f90XEsxHx7O6u/8VQAICHz/3OEzdv3TqmIwUAPAj3O0/cWWOeAID7YW8CRcSSpF+Q9IGU0qakn5T0ZklPa29n/8fv9f+llD6UUnompfTMcOh9szwA4OFzGPPExQsXju14AQDH6zDmidXzzBMAcD+sTaCIKLU3YP/jlNIvSlJK6XpKqU4pNZJ+StK7ju4wAQAnGfMEAOAgzBMAcDI4vx0sJP20pC+klP7WXT9/9K7Y90r63OEfHgDgpGOeAAAchHkCAE4O57eDfbuk75P0WxHxmf2f/Q1J74mIpyUlSc9L+oEjOUIAwEnHPAEAOAjzBACcEM5vB/u0pLjHX328bWXT6VRff+nFubmUkl1mXdd23Y5ZNfPKm3k5SZrNvLqnU6/MyXhs152qxsrl5tdDZXGvpnCPemv/HkbmlRm5mYvczHnlSVJqzPNx2653WyRJYd6bPHP2dKWs9HKLi8tWTpIG/YGV6/V6Vs49Z0na3fG+cD5rvPvdNN7NyXLvOkpev2nTHk+aw5wnkrx7UJljvyRNp5WVm5hj8Kzy6q5ajIONOcaE2TVaDDGq73nr7lG3mcsOOdeGe8XtZYZ/C9tETYd/fdxrnjKvoYWZ28t6c7PMOTy1GDPdunsDby4bLHjfcdnre3OeJBW9cm7GXS+dRIc5T4Sk0hiGh+a6SJLOLCxZucWBd+/Lwqt7sLhg5SRpcPaMleud8c6lf2bFrnu46tWdGe1YkhpzTb9lPkNJ0k5lPm+Zw9YkvJl01GI9Mktedpa8utuMCIW5gFhY8tb/A3MdvHvjtpWTpLSxZeXyoTe2Vi0u0PZoZOUm5lSWel4wH/jzaDmcn3XXia1+OxgAAAAAAAAeTmwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdEBxnJWVZanHLl+em5vOZnaZu+OxlRuNRlYuy719saIsrZwkFYV3ma9fv27lbty4Yde9s7Nt5fLcO0Y3556zJGVZbubCykV4OTXJy7Uo084V3jlLUmTmXq152qmprdzCcOAVKCnLFqxcr9ezcnn4+9PT8cQLmvc7z717k7v3RV67SMlvj6daSmqaZm6sqiq7yPHEmyfc+WQym1q5We0fo9crpczs6HUyBwRJ+fzLLUlyh1a36uYIXoZKbuUPtLu5lfsH6Q4fbqvIzQKzFuNWlryGVhn9X5JmldtrpNos051vM3OesDuNHnCTfMiEQoOYfw8mlTm4Saq3d6xcs7Nr5dLEmyfyFu14aeCtod7wTW+ycgvnV+26t9x5zx3/zeetqPxnwsLMTsJrF7vmmBWNf2FgiKYAAA8uSURBVIx57WWzxls/ZG3menNt3R946/9hr++Vd96KSZKavvnsMfLWaxtj/95sVt5z+Ib5bH1z446Ve/TispWTJOVL8zNmk+CdQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAD+//buLtSy867j+O+/195nn5mTyZx5MwnJVFvpRUE0lVACipRQpXrTClUsCBGEeqFQ8UbxxlYUVHy7q1RaiKCmoa22eGUuUtSbtGmb9C2YpiVqzJBJJhnnvO+X9ffi7IEhnHP2b/Xsc/Z6+X5gmHP2/M96nv/zrPX813lmvwDoADaBAAAAAAAAOqB/qo0VfV26eGluXGbax5yWpRU3KadWXL+/+CHZHe1ZcdeuXbPibt26Zbe9sbFpxa2vX7DiVldXFxonSRFhxZXuXE8mZsNemCT1zD72et6+apVzfGd3x4ozu6iV4YoXV+Fa6Bde7KBXWHEVpsY/L8ZjK246Nc+fCnOYOb+P06m3RrVehHUdueuGJE2ni1073Lmqcp27isK7hvoD//odmGuCO+aLjqvCvo7suVl8HyWv7SrnT1l6se6Y98w1vSj8/0u05zu8Y07NnCXJDS36AyuuZ8ZNKvRxbzx//SlPYE1popA0MIZixbwnk6Sz5jo4XDHvd8zz/ebNG1acJO0U3vp2zzveZsWdDe/eX5KKvjeWI/Meanc0suLK8M/5YuCN+erAm2uZvztub+56x5M02fN+JxyPvfEZuLlIGg6HXtsjL59d8756zbsF21d6c7i1sW3FTXe88ZakwrxmFx2XWeF+1hhL94rhmUAAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAd0D/NxsostbOzs9iDRiz0cMWKuS/W89vN3bTiNjc2rbjt7W277cFgYMWtrZ214s6e9eLcdiUp0xuf6XRqxYV7TpReuyfBzUWSxpOJFZdZWnHuzm8h/xwvC6/tLL04meeEJLuXPfOa7ZnnT8rvo5P2YleyBstUaQyYu25IfpkoisKKc9e3Kuugy817PB77xzTjip63evT75viYcZLUC6/tMOfwJLhzs+i4fd7a6tbHovDmJircC8msUe4xB8MVu+nh6qoVtzIcWnH9/uJvn511r0ptbLVMTcv591Hj0cg+5PbWlte0eU+2OvSuoal57UrSZHfPinvxhResuNdu3bTbvvvyRStuZW3NijtnXpOj9O+XR8Y5IUlbm97vojn2zp/ViV9vi/BqVA69cXRroyQVU/P+ds/LO8betTDZ9Men3PB+x5286Z27xdau3fY5c1ukP/HOs/G2d57tVOjj1ur8uSmnXp3gmUAAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAd0D/NxqbjiW689vrcuCj8val+30thOBxacYV5PIUXJknbW1tW3N7enhU3nU7ttns9byzdcXTjIvwBysyFHrNnxpXy2pX8PrrccZSkwWBgxZXmeTHa3bHiCvPckaSBmY977k6nE7vtcur1M/qFFefPjXc8Scpi/vlT5Zpps+l0qlu3bs2N2xt566Xkz+n6+roVd/Xq1YUeT5ImE++cL8vSiquyZrltj8wxH43GVtzunrcWSdJ0Yq4d7viU5vhUuCzDDD6Ja33RNSrM9b9SJrHYPl44f96OvXzlihV37tw5K86+pyz8OuGdF9SJ29K41gcVxv/u1TNm3FkvbsU7Xkz8WrY79ta3yY3/s+I29vx7rd3X3rTi3HstmWtMr+ffL7u/P5bmZeSu1Wf7/v1y9LzxKcxccuKvq6VZR3PsnRfTkXnvsLNrxUnSeNeLLXe966bYGdlt96beWA7Tm5sVt456l7Uk717IvR/gmUAAAAAAAAAdMHcTKCJWI+LLEfFcRHw7Ij4+e/ztEfF0RHw3Ij4TESsn310AQN1QJwAAR6FOAEB9OM8E2pP0SGb+hKQHJb0/Ih6W9KeS/ioz3ynpTUm/fnLdBADUGHUCAHAU6gQA1MTcTaDctzn7djD7k5IekfTZ2eOPSfrgifQQAFBr1AkAwFGoEwBQH9Z7AkVEERHPSrou6UlJ35N0MzNvvyPUy5LuP+RnPxIRz0TEMzt7/htDAQCaY1F14saNN06nwwCAU7W4OjH/Q2YAAIezNoEyc5qZD0p6QNJ7JL3roLBDfvaTmflQZj50Zrj6g/cUAFBbi6oTly5dPMluAgCWZHF14vJJdhMAWq/Sp4Nl5k1JX5L0sKT1iLj9uX0PSHplsV0DADQNdQIAcBTqBAAsl/PpYFciYn329RlJ75P0vKSnJH1oFvaopC+cVCcBAPVFnQAAHIU6AQD10Z8fovskPRYRhfY3jZ7IzH+JiO9Iejwi/kjS1yV96gT7CQCoL+oEAOAo1AkAqIm5m0CZ+Q1J7z7g8e9r//W8AIAOo04AAI5CnQCA+nCeCbQw07LUxsbG3LiiX9jH7A8GVtze3p4Vt7G1OT9I0mQ6teIkaWdnx4pbHQ6tuEsXL9ltT0uvn0XhvT1UaR4v88D39TvkmOVijxlhhfX7iz/93T6G2UfJHx83bjqZzA+SNDHjqhwzB96YZ4XrSwse817PuxaqzKF1XlQ4Xptleude0fPrxPnz5624tbW7rLh7773XiptWOI+rnE8Odz2QpPF4bMVtb29bcbu73ieBunVZkkajkRXn5uKub1XmcNG1rEod9WO9uNI8XFlhfNwxLwrv2j5z5ozd9oULF6y4ixe9N6ZfWVmx4qrMoTM+VY7XZmVIo2L+WJRDv04M7l6z4i7e470p9aTnzdXG1pYVJ0k75tqqnldPcstbLyUpN7w12Fxi1DNrXvqlzL4fLCsddL4q9yO9WOw9ZpU12I415zDNQPdakCQVXt4r5u//xV3efZ0k9c6ds+KumB9gcuVu797zTIV9D6X/u9k8ld4YGgAAAAAAAM3EJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB3AJhAAAAAAAEAHsAkEAAAAAADQAWwCAQAAAAAAdACbQAAAAAAAAB0QmXl6jUW8Jum/3vLwZUmvn1onTlabcpHalQ+51Feb8jlOLj+cmVcW2Zkmok40TpvyIZd6alMuEnXi2KgTjdKmXKR25UMu9XXideJUN4EO7EDEM5n50FI7sSBtykVqVz7kUl9tyqdNudRJm8a1TblI7cqHXOqpTblI7cunLto0ruRSX23Kh1zq6zTy4eVgAAAAAAAAHcAmEAAAAAAAQAfUYRPok8vuwAK1KRepXfmQS321KZ825VInbRrXNuUitSsfcqmnNuUitS+fumjTuJJLfbUpH3KprxPPZ+nvCQQAAAAAAICTV4dnAgEAAAAAAOCELXUTKCLeHxH/GREvRsTvLbMvxxURL0XENyPi2Yh4Ztn9qSIiPh0R1yPiW3c8djEinoyI787+vrDMPlZxSD4fi4j/nc3PsxHxC8vsoysirkbEUxHxfER8OyI+Onu8cfNzRC6Nm5uIWI2IL0fEc7NcPj57/O0R8fRsXj4TESvL7muTtalGSNSJOqFO1BN1AlVRJ+qjTXWCGlFf1IkFtb2sl4NFRCHpBUk/K+llSV+R9OHM/M5SOnRMEfGSpIcy8/Vl96WqiPgZSZuS/i4zf2z22J9JeiMz/2RWVC9k5u8us5+uQ/L5mKTNzPzzZfatqoi4T9J9mfm1iDgn6auSPijp19Sw+Tkil19Ww+YmIkLSWmZuRsRA0n9I+qik35H0+cx8PCL+RtJzmfmJZfa1qdpWIyTqRJ1QJ+qJOoEqqBP10qY6QY2oL+rEYizzmUDvkfRiZn4/M0eSHpf0gSX2p7My898kvfGWhz8g6bHZ149p/+JqhEPyaaTMvJaZX5t9vSHpeUn3q4Hzc0QujZP7NmffDmZ/UtIjkj47e7wR81Jj1IgaoU7UF3WinqgTp4I6USNtqhPUiPqiTizGMjeB7pf0P3d8/7IaOoEzKelfI+KrEfGRZXdmAe7JzGvS/sUm6YeW3J9F+K2I+MbsKZ6NeMrjnSLiRyS9W9LTavj8vCUXqYFzExFFRDwr6bqkJyV9T9LNzJzMQpq+pi1b22qERJ1ogsatRXeiTtQLdeLEUSfqr9Hr0AEatw7dqU01QqJOHMcyN4HigMea/FFlP5WZPynp5yX95uxphKiPT0j6UUkPSrom6S+W251qIuIuSZ+T9NuZeWvZ/TmOA3Jp5Nxk5jQzH5T0gPb/N/JdB4Wdbq9apW01QqJO1F0j16LbqBP1Q504cdQJnKZGrkO3talGSNSJ41rmJtDLkq7e8f0Dkl5ZUl+OLTNfmf19XdI/aX8Sm+zV2Wsub7/28vqS+3Msmfnq7CIrJf2tGjQ/s9eIfk7S32fm52cPN3J+DsqlyXMjSZl5U9KXJD0saT0i+rN/avSaVgOtqhESdaLumrwWUSfqjTpxYqgT9dfIdeggTV6H2lQjJOrEIixzE+grkt45e/frFUm/IumLS+zPDywi1mZvTKWIWJP0c5K+dfRP1d4XJT06+/pRSV9YYl+O7fYiN/OLasj8zN4w7FOSns/Mv7zjnxo3P4fl0sS5iYgrEbE++/qMpPdp/zXJT0n60CysEfNSY62pERJ1ogmauBZJ1Im6ok6cCupE/TVuHTpME9chqV01QqJOLKztXNKng0lS7H90219LKiR9OjP/eGmdOYaIeIf2d+slqS/pH5qUS0T8o6T3Sros6VVJfyDpnyU9Ieltkv5b0i9lZiPeIO2QfN6r/acHpqSXJP3G7dfB1llE/LSkf5f0TUnl7OHf1/5rXxs1P0fk8mE1bG4i4se1/0ZthfY305/IzD+crQWPS7oo6euSfjUz95bX02ZrS42QqBN1Q52oJ+oEqqJO1Eeb6gQ1or6oEwtqe5mbQAAAAAAAADgdy3w5GAAAAAAAAE4Jm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AFsAgEAAAAAAHQAm0AAAAAAAAAdwCYQAAAAAABAB7AJBAAAAAAA0AH/DzgjqhFLq4O9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c14cd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This function should take in an RGB image and return a new, standardized version\n",
    "def standardize_input(image):\n",
    "    ## TODO: Resize image and pre-process so that all \"standard\" images are the same size  \n",
    "    box, angle, sobel = getTrafficLightAngle(image)\n",
    "    \n",
    "    # create rectangluar mask from edge detection\n",
    "    mask = rectanglemask(sobel, box)\n",
    "    mask = cv2.resize(mask, (image.shape[1], image.shape[0])) \n",
    "    mask_img = np.copy(image)\n",
    "    #mask image\n",
    "    mask_img = cv2.bitwise_and(image, mask_img,mask = mask)\n",
    "\n",
    "    # rotate image by skew\n",
    "    cnt = rotate(mask, angle)\n",
    "    # get bounding box\n",
    "    x,y,w,h = cv2.boundingRect(cnt)\n",
    "    #rotate masked image & crop\n",
    "    rotated = rotate(mask_img, angle)\n",
    "    cropped = rotated[y:y+h,x:x+w]\n",
    "    \n",
    "    #resize to 32 X 32\n",
    "    standard_im = cv2.resize(cropped, (32, 32))\n",
    "    \n",
    "    return standard_im\n",
    "    \n",
    "\n",
    "#original red green and blue image\n",
    "random_red, _, index1 = getImage(color='red', Random=True)\n",
    "random_yellow, _, index2 = getImage(color='yellow', Random=True)\n",
    "random_green, _, index3 = getImage(color='green', Random=True)\n",
    "\n",
    "# show red, yellow, green images\n",
    "f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20,10))\n",
    "ax1.set_title('red')\n",
    "ax1.imshow(random_red)\n",
    "ax2.set_title('yellow')\n",
    "ax2.imshow(random_yellow)\n",
    "ax3.set_title('green')\n",
    "ax3.imshow(random_green)\n",
    "plt.show()\n",
    "    \n",
    "#original red green and blue image\n",
    "std_red = standardize_input(random_red)\n",
    "std_yellow = standardize_input(random_yellow)\n",
    "std_green = standardize_input(random_green)\n",
    "\n",
    "# show red, yellow, green images\n",
    "f2, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20,10))\n",
    "ax1.set_title('red')\n",
    "ax1.imshow(std_red)\n",
    "ax2.set_title('yellow')\n",
    "ax2.imshow(std_yellow)\n",
    "ax3.set_title('green')\n",
    "ax3.imshow(std_green)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Standardize the output\n",
    "\n",
    "With each loaded image, we also specify the expected output. For this, we use **one-hot encoding**.\n",
    "\n",
    "* One-hot encode the labels. To do this, create an array of zeros representing each class of traffic light (red, yellow, green), and set the index of the expected class number to 1. \n",
    "\n",
    "Since we have three classes (red, yellow, and green), we have imposed an order of: [red value, yellow value, green value]. To one-hot encode, say, a yellow light, we would first initialize an array to [0, 0, 0] and change the middle value (the yellow value) to 1: [0, 1, 0].\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='task3'></a>\n",
    "### (IMPLEMENTATION): Implement one-hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "## TODO: One hot encode an image label\n",
    "## Given a label - \"red\", \"green\", or \"yellow\" - return a one-hot encoded label\n",
    "\n",
    "# Examples: \n",
    "# one_hot_encode(\"red\") should return: [1, 0, 0]\n",
    "# one_hot_encode(\"yellow\") should return: [0, 1, 0]\n",
    "# one_hot_encode(\"green\") should return: [0, 0, 1]\n",
    "\n",
    "def one_hot_encode(label):\n",
    "    \n",
    "    ## TODO: Create a one-hot encoded label that works for all classes of traffic lights\n",
    "    one_hot_encoded = [0, 0, 0]\n",
    "    # check whether color is red green of yellow\n",
    "    acceptable_colors = tuple(['red', 'yellow', 'green'])\n",
    "    if (not label in acceptable_colors):\n",
    "            raise ValueError('label: {} is not an acceptable color'.format(label))\n",
    "    if label == 'red':\n",
    "        one_hot_encoded[0] = 1\n",
    "    if label == 'yellow':\n",
    "        one_hot_encoded[1] = 1\n",
    "    if label == 'green':\n",
    "        one_hot_encoded[2] = 1\n",
    "    \n",
    "    return one_hot_encoded\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Testing as you Code\n",
    "\n",
    "After programming a function like this, it's a good idea to test it, and see if it produces the expected output. **In general, it's good practice to test code in small, functional pieces, after you write it**. This way, you can make sure that your code is correct as you continue to build a classifier, and you can identify any errors early on so that they don't compound.\n",
    "\n",
    "All test code can be found in the file `test_functions.py`. You are encouraged to look through that code and add your own testing code if you find it useful!\n",
    "\n",
    "One test function you'll find is: `test_one_hot(self, one_hot_function)` which takes in one argument, a one_hot_encode function, and tests its functionality. If your one_hot_label code does not work as expected, this test will print ot an error message that will tell you a bit about why your code failed. Once your code works, this should print out TEST PASSED."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "**<span style=\"color: green;\">TEST PASSED</span>**"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Importing the tests\n",
    "import test_functions\n",
    "tests = test_functions.Tests()\n",
    "\n",
    "# Test for one_hot_encode function\n",
    "tests.test_one_hot(one_hot_encode)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Construct a `STANDARDIZED_LIST` of input images and output labels.\n",
    "\n",
    "This function takes in a list of image-label pairs and outputs a **standardized** list of resized images and one-hot encoded labels.\n",
    "\n",
    "This uses the functions you defined above to standardize the input and output, so those functions must be complete for this standardization to work!\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def standardize(image_list):\n",
    "    \n",
    "    # Empty image data array\n",
    "    standard_list = []\n",
    "\n",
    "    # Iterate through all the image-label pairs\n",
    "    for item in image_list:\n",
    "        image = item[0]\n",
    "        label = item[1]\n",
    "\n",
    "        # Standardize the image\n",
    "        standardized_im = standardize_input(image)\n",
    "\n",
    "        # One-hot encode the label\n",
    "        one_hot_label = one_hot_encode(label)    \n",
    "\n",
    "        # Append the image, and it's one hot encoded label to the full, processed list of image data \n",
    "        standard_list.append((standardized_im, one_hot_label))\n",
    "        \n",
    "    return standard_list\n",
    "\n",
    "# Standardize all training images\n",
    "STANDARDIZED_LIST = standardize(IMAGE_LIST)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize the standardized data\n",
    "\n",
    "Display a standardized image from STANDARDIZED_LIST and compare it with a non-standardized image from IMAGE_LIST. Note that their sizes and appearance are different!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f9b3bebeda0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3e474080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## TODO: Display a standardized image and its label\n",
    "total_num_of_images = len(STANDARDIZED_LIST)\n",
    "image_index = random.randint(0, total_num_of_images-1)\n",
    "\n",
    "f2, (ax1, ax2) = plt.subplots(1, 2, figsize=(20,10))\n",
    "ax1.set_title(STANDARDIZED_LIST[image_index][1])\n",
    "ax1.imshow(STANDARDIZED_LIST[image_index][0])\n",
    "ax2.set_title(IMAGE_LIST[image_index][1])\n",
    "ax2.imshow(IMAGE_LIST[image_index][0])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Feature Extraction\n",
    "\n",
    "You'll be using what you now about color spaces, shape analysis, and feature construction to create features that help distinguish and classify the three types of traffic light images.\n",
    "\n",
    "You'll be tasked with creating **one feature** at a minimum (with the option to create more). The required feature is **a brightness feature using HSV color space**:\n",
    "\n",
    "1. A brightness feature.\n",
    "    - Using HSV color space, create a feature that helps you identify the 3 different classes of traffic light.\n",
    "    - You'll be asked some questions about what methods you tried to locate this traffic light, so, as you progress through this notebook, always be thinking about your approach: what works and what doesn't?\n",
    "\n",
    "2. (Optional): Create more features! \n",
    "\n",
    "Any more features that you create are up to you and should improve the accuracy of your traffic light classification algorithm! One thing to note is that, to pass this project you must **never classify a red light as a green light** because this creates a serious safety risk for a self-driving car. To avoid this misclassification, you might consider adding another feature that specifically distinguishes between red and green lights.\n",
    "\n",
    "These features will be combined near the end of his notebook to form a complete classification algorithm."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating a brightness feature \n",
    "\n",
    "There are a number of ways to create a brightness feature that will help you characterize images of traffic lights, and it will be up to you to decide on the best procedure to complete this step. You should visualize and test your code as you go.\n",
    "\n",
    "Pictured below is a sample pipeline for creating a brightness feature (from left to right: standardized image, HSV color-masked image, cropped image, brightness feature):\n",
    "\n",
    "<img src=\"images/feature_ext_steps.png\" width=\"70%\" height=\"70%\">\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RGB to HSV conversion\n",
    "\n",
    "Below, a test image is converted from RGB to HSV colorspace and each component is displayed in an image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label [red, yellow, green]: [1, 0, 0]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f9b3bdc27f0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c3754e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3bfe69b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Convert and image to HSV colorspace\n",
    "# Visualize the individual color channels\n",
    "\n",
    "image_num = random.randint(0, total_num_of_images-1)\n",
    "test_im = STANDARDIZED_LIST[image_num][0]\n",
    "test_label = STANDARDIZED_LIST[image_num][1]\n",
    "\n",
    "# Convert to HSV\n",
    "hsv = cv2.cvtColor(test_im, cv2.COLOR_RGB2HSV)\n",
    "\n",
    "# Print image label\n",
    "print('Label [red, yellow, green]: ' + str(test_label))\n",
    "\n",
    "# HSV channels\n",
    "h = hsv[:,:,0]\n",
    "s = hsv[:,:,1]\n",
    "v = hsv[:,:,2]\n",
    "\n",
    "# Plot the original image and the three channels\n",
    "f, (ax1, ax2, ax3, ax4) = plt.subplots(1, 4, figsize=(20,10))\n",
    "ax1.set_title('Standardized image')\n",
    "ax1.imshow(test_im)\n",
    "ax2.set_title('H channel')\n",
    "ax2.imshow(h, cmap='gray')\n",
    "ax3.set_title('S channel')\n",
    "ax3.imshow(s, cmap='gray')\n",
    "ax4.set_title('V channel')\n",
    "ax4.imshow(v, cmap='gray')\n",
    "\n",
    "# rgb channels\n",
    "r = test_im[:,:,0]\n",
    "g = test_im[:,:,1]\n",
    "b = test_im[:,:,2]\n",
    "\n",
    "# Plot the original image and the three channels\n",
    "f, (ax1, ax2, ax3, ax4) = plt.subplots(1, 4, figsize=(20,10))\n",
    "ax1.set_title('Standardized image')\n",
    "ax1.imshow(test_im)\n",
    "ax2.set_title('R channel')\n",
    "ax2.imshow(r, cmap='gray')\n",
    "ax3.set_title('G channel')\n",
    "ax3.imshow(g, cmap='gray')\n",
    "ax4.set_title('B channel')\n",
    "ax4.imshow(b, cmap='gray')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='task7'></a>\n",
    "### (IMPLEMENTATION): Create a brightness feature that uses HSV color space\n",
    "\n",
    "Write a function that takes in an RGB image and returns a 1D feature vector and/or single value that will help classify an image of a traffic light. The only requirement is that this function should apply an HSV colorspace transformation, the rest is up to you. \n",
    "\n",
    "From this feature, you should be able to estimate an image's label and classify it as either a red, green, or yellow traffic light. You may also define helper functions if they simplify your code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Helper functions\n",
    "#### List multiplication\n",
    "\n",
    "def matrix_scalar_mul(matrix, scalar):\n",
    "    new = []\n",
    "    for i in range(len(matrix)):\n",
    "        new.append(matrix[i] * scalar)\n",
    "    return new\n",
    "\n",
    "\n",
    "def matrix_multiplication(matrixA, matrixB):\n",
    "    product = []\n",
    "    \n",
    "    if len(matrixA) != len(matrixB):\n",
    "        raise ValueError('list must be the same size, A:', len(matrixA), 'B:', len(matrixB))\n",
    "\n",
    "    for i in range(len(matrixA)):\n",
    "        product.append(matrixA[i] * matrixB[i])\n",
    "\n",
    "    return product\n",
    "\n",
    "########################\n",
    "\n",
    "\n",
    "# return list index which has a maximum value\n",
    "def max_idx(yvals, ranges):\n",
    "    mx = 0\n",
    "    j = 0\n",
    "    for i in ranges:\n",
    "        if yvals[i] > mx:\n",
    "            mx = yvals[i]\n",
    "            j = i\n",
    "    return j\n",
    "\n",
    "# return list index which has a minimum value\n",
    "def min_idx(yvals, ranges):\n",
    "    mn = max(yvals)\n",
    "    j = 0\n",
    "    for i in ranges:\n",
    "        if yvals[i] < mn and yvals[i] > 0:\n",
    "            mn = yvals[i]\n",
    "            j = i\n",
    "    return j\n",
    "\n",
    "# return the top x indicies from a list \n",
    "def max_idx_rank(yvals):\n",
    "    indicies = set(range(len(yvals)))\n",
    "    # creat a list to append max bins\n",
    "    max_list = []\n",
    "    # create set to perform set operations\n",
    "    max_set = set()\n",
    "    intersect = indicies - max_set\n",
    "    \n",
    "    # rank first 8 bins\n",
    "    for i in range(0, 8):\n",
    "        # append next maximum value\n",
    "        max_list.append(max_idx(yvals, intersect))\n",
    "        # add value to set list\n",
    "        max_set.add(max_list[-1])\n",
    "        # remove bin from bin list\n",
    "        intersect = indicies - max_set\n",
    "    \n",
    "    return max_list\n",
    "\n",
    "# function to determine if the distribution is bimodal or normal\n",
    "def is_bimodal(max_list, values):\n",
    "    difference = []\n",
    "    for i in range(len(max_list)-1):\n",
    "        for j in range(max_list[i], max_list[i+1]):\n",
    "            if values[j] == 0:\n",
    "                return True\n",
    "    \n",
    "    return False\n",
    "\n",
    "# # find the mean index\n",
    "# def mean_bin(values):\n",
    "#     #change zeros to nan for mean\n",
    "#     mean_vals = [np.nan if x == 0 else x for x in values]\n",
    "#     mean = np.nanmean(mean_vals)\n",
    "#     print(mean)\n",
    "#     difference = []\n",
    "#     for i in range(len(values)):\n",
    "#         difference.append(abs(values[i] - mean))\n",
    "        \n",
    "#     mean_i = min_idx(difference, range(len(difference)))\n",
    "    \n",
    "#     return mean_i\n",
    "    \n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "## TODO: Create a brightness feature that takes in an RGB image and outputs a feature vector and/or value\n",
    "## This feature should use HSV colorspace values\n",
    "\n",
    "\n",
    "def color_channels(image, channel, n_bins = 24):\n",
    "    if channel == \"hsv\":\n",
    "        # Convert absimage to HSV color space\n",
    "        hsv = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)\n",
    "        # Create color channel histograms\n",
    "        h_hist = np.histogram(hsv[:,:,0], bins=n_bins, range=(0, 180))\n",
    "        s_hist = np.histogram(hsv[:,:,1], bins=n_bins, range=(0, 256))\n",
    "        v_hist = np.histogram(hsv[:,:,2], bins=n_bins, range=(0, 256))\n",
    "        \n",
    "        return h_hist, s_hist, v_hist \n",
    "    else:\n",
    "        rgb = np.copy(image)\n",
    "        # Create color channel histograms\n",
    "        r_hist = np.histogram(rgb[:,:,0], bins=n_bins, range=(0, 256))\n",
    "        g_hist = np.histogram(rgb[:,:,1], bins=n_bins, range=(0, 256))\n",
    "        b_hist = np.histogram(rgb[:,:,2], bins=n_bins, range=(0, 256))\n",
    "        \n",
    "        return r_hist, g_hist, b_hist\n",
    "    \n",
    "def color_isolate(image, channel):\n",
    "    if channel == \"hsv\":\n",
    "        # Convert absimage to HSV color space\n",
    "        hsv = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)\n",
    "        # Create color channels\n",
    "        h = hsv[:,:,0]\n",
    "        s = hsv[:,:,1]\n",
    "        v = hsv[:,:,2]\n",
    "        \n",
    "        return h, s, v \n",
    "    else:\n",
    "        rgb = np.copy(image)\n",
    "        # Create color channels\n",
    "        r = rgb[:,:,0]\n",
    "        g = rgb[:,:,1]\n",
    "        b = rgb[:,:,2]\n",
    "        \n",
    "        return r, g, b \n",
    "\n",
    "# plot histograms of rbg, or hsv space along the y-axis of the image\n",
    "def yaxis_hists(rgb_image, channel):\n",
    "    # seperate image out into different channels of color space\n",
    "    c1, c2, c3 = color_isolate(rgb_image, 'hsv')\n",
    "    # Sum components over all coloumns for each row (axis = 1)\n",
    "    hist_sum = []\n",
    "    c1_sum = np.sum(c1[:,:], axis=1)\n",
    "    c2_sum = np.sum(c2[:,:], axis=1)\n",
    "    c3_sum = np.sum(c3[:,:], axis=1)\n",
    "    \n",
    "    #get baselines\n",
    "    base1 = np.median(c1_sum)\n",
    "    base2 = np.median(c2_sum)\n",
    "    base3 = np.median(c3_sum)\n",
    "    \n",
    "    # split histrogram around the median\n",
    "    c1_norm = matrix_scalar_mul((c1_sum - base1).tolist(), -1)\n",
    "    c2_norm = (c2_sum - base2).tolist()\n",
    "    c3_norm = (c3_sum - base3).tolist()\n",
    "    \n",
    "    # get rid of negative values\n",
    "    #np.nan\n",
    "    c1_norm = [0 if x < 0 else x for x in c1_norm]\n",
    "    c2_norm = [0 if x < 0 else x for x in c2_norm]\n",
    "    c3_norm = [0 if x < 0 else x for x in c3_norm]\n",
    "    \n",
    "    # package as 2D list\n",
    "    hist_vals = []\n",
    "    hist_vals.append(c1_norm)\n",
    "    hist_vals.append(c2_norm)\n",
    "    hist_vals.append(c3_norm)\n",
    "    \n",
    "    # get bins\n",
    "    bin_edges = range(rgb_image.shape[0])\n",
    "\n",
    "    return bin_edges, hist_vals\n",
    "\n",
    "def value_hists(rgb_image, channel):\n",
    "    c1, c2, c3 = color_channels(rgb_image, channel)\n",
    "    \n",
    "    hist_vals = []\n",
    "    hist_vals.append(c1[0])\n",
    "    hist_vals.append(c2[0])\n",
    "    hist_vals.append(c3[0])\n",
    "    \n",
    "    # Generating bin centers\n",
    "    bin_edges = c1[1]\n",
    "    bin_centers = (bin_edges[1:]  + bin_edges[0:len(bin_edges)-1])/2\n",
    "    \n",
    "    return bin_centers, hist_vals\n",
    "    \n",
    "\n",
    "def plotHist(bins, values, thickness, channel, yLbl, yLim = False):\n",
    "    if channel == 'hsv':\n",
    "        names = ('hue','saturation','value')\n",
    "    else:\n",
    "        names = ('red','green','blue')\n",
    "        \n",
    "    plt.rcdefaults()\n",
    "    f1, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20,10))\n",
    "    \n",
    "    #first channel\n",
    "    hist1 = ax1.barh(bins, values[0], align=\"center\", height=thickness)\n",
    "    # get mean and bimodal bool\n",
    "    max_list = max_idx_rank(values[0])\n",
    "    mean = max_list[0]\n",
    "    bimodal = is_bimodal(max_list, values[0])\n",
    "    \n",
    "    ax1.invert_yaxis()\n",
    "    ax1.set_ylabel(yLbl)\n",
    "    ax1.set_xlabel(\"amount\")\n",
    "    ax1.set_title(\"%s\"%names[0])\n",
    "    ax1.set_title(\"%s :mu = %i, bi = %r\" %(names[0],mean, bimodal))\n",
    "    if yLim:\n",
    "        ax1.set_xlim(0, 256)\n",
    "            \n",
    "    #second channel\n",
    "    hist2 = ax2.barh(bins, values[2], align=\"center\", height=thickness)\n",
    "    # get mean and bimodal bool\n",
    "    max_list2 = max_idx_rank(values[2])\n",
    "    mean2 = max_list2[0]\n",
    "    bimodal2 = is_bimodal(max_list2, values[2])\n",
    "    \n",
    "    ax2.invert_yaxis()\n",
    "    ax2.set_ylabel(yLbl)\n",
    "    ax2.set_xlabel(\"amount\")\n",
    "    ax2.set_title(\"%s :mu = %i, bi = %r\" %(names[2],mean2, bimodal2))\n",
    "    if yLim:\n",
    "        ax2.set_xlim(0, 256)\n",
    "        \n",
    "    #third channel\n",
    "    sat_val = matrix_multiplication(values[2], values[0])\n",
    "    hist3 = ax3.barh(bins, sat_val, align=\"center\", height=thickness)\n",
    "    # get mean and bimodal bool\n",
    "    max_list3 = max_idx_rank(sat_val)\n",
    "    mean3 = max_list3[0]\n",
    "    bimodal3 = is_bimodal(max_list3, sat_val)\n",
    "    \n",
    "    ax3.invert_yaxis()\n",
    "    ax3.set_ylabel(yLbl)\n",
    "    ax3.set_xlabel(\"amount\")\n",
    "    ax3.set_title(\"%s :mu = %i, bi = %r\" %('hue+val',mean3, bimodal3))\n",
    "    if yLim:\n",
    "        ax3.set_xlim(0, 256)\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3bcae470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 0, 0]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f9b3bf67da0>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3bd68a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def feature_value(rgb_image, plot = False):\n",
    "    ## TODO: Convert absimage to HSV color space\n",
    "    ## TODO: Create and return a feature value and/or vector\n",
    "    \n",
    "    # calculate HSVspace over the height on the traffic light\n",
    "    bins, values = yaxis_hists(rgb_image, 'hsv')\n",
    "    \n",
    "    # for testing purposes\n",
    "    if plot == True:\n",
    "        plotHist(bins, values, 0.8, 'hsv', 'y-dist')\n",
    "    \n",
    "    return values[2]\n",
    "\n",
    "# image, label, image_num = getImage(color='yellow', Random=True, showImg=False)\n",
    "# #image_num = random.randint(0, total_num_of_images-1)\n",
    "# test_im = STANDARDIZED_LIST[image_num][0]\n",
    "# test_label = STANDARDIZED_LIST[image_num][1]\n",
    "\n",
    "feature = feature_value(test_im, True)\n",
    "print(test_label)\n",
    "plt.imshow(test_im)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## (Optional) Create more features to help accurately label the traffic light images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def feature_valueXHue(rgb_image, plot = False):\n",
    "    ## TODO: Convert absimage to HSV color space\n",
    "    ## TODO: Create and return a feature value and/or vector\n",
    "    \n",
    "    # calculate HSVspace over the height on the traffic light\n",
    "    bins, values = yaxis_hists(rgb_image, 'hsv')\n",
    "    hue_val = matrix_multiplication(values[2], values[0])\n",
    "    \n",
    "    return hue_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 0, 0]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c257c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3be9dcc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# (Optional) Add more image analysis and create more features\n",
    "#####\n",
    "# This feature will look at the hue of the image. If the max hue is within a \n",
    "# color range it will return that color.\n",
    "#####\n",
    "# hue ranges\n",
    "    # * red = (0, 10) || (160, 180)\n",
    "    # * yellow = (90, 130) \n",
    "    # * green = (80, 140) \n",
    "    \n",
    "def feature_rgb(rgb_image, plot = False):\n",
    "    ## TODO: Convert absimage to HSV color space\n",
    "    ## TODO: Create and return a feature value and/or vector\n",
    "    \n",
    "#     # calculate hsvspace hue over the whole image\n",
    "#     bins, values = value_hists(rgb_image, 'rgb')\n",
    "#     if plot:\n",
    "#         plotHist(bins, values, 6, 'rgb', 'value', False)\n",
    "    # Convert absimage to HSV color space\n",
    "    hsv = cv2.cvtColor(rgb_image, cv2.COLOR_RGB2HSV)\n",
    "    \n",
    "    \n",
    "    image = np.copy(hsv)\n",
    "    # green\n",
    "    lower_green = np.array([80,120,80]) \n",
    "    upper_green = np.array([140,255,255])\n",
    "    #yellow\n",
    "    lower_yellow = np.array([0,100,120]) \n",
    "    upper_yellow = np.array([80,255,255])\n",
    "    #red\n",
    "    lower_red = np.array([140,50,20]) \n",
    "    upper_red = np.array([180,255,255])\n",
    "    \n",
    "    mask_yellow = cv2.inRange(hsv, lower_yellow, upper_yellow)\n",
    "    mask_red = cv2.inRange(hsv, lower_red, upper_red)\n",
    "    mask_green = cv2.inRange(hsv, lower_green, upper_green)\n",
    "    \n",
    "    masked_image = np.zeros((image.shape[1],image.shape[0],3), np.uint8)\n",
    "    masked_image[mask_red != 0] = [255, 0, 0]\n",
    "    masked_image[mask_green != 0] = [0, 255, 0]\n",
    "    masked_image[mask_yellow != 0] = [0, 0, 255]\n",
    "        \n",
    "    return masked_image\n",
    "\n",
    "# #image, label, image_num = getImage(color='red', Random=True, showImg=False)\n",
    "# #image_num = random.randint(0, total_num_of_images-1)\n",
    "# image_num = 757\n",
    "# test_im = STANDARDIZED_LIST[image_num][0]\n",
    "# test_label = STANDARDIZED_LIST[image_num][1]\n",
    "\n",
    "feature = feature_rgb(test_im, True)\n",
    "print(test_label)\n",
    "plt.imshow(feature)\n",
    "plt.show()\n",
    "plt.imshow(test_im)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (QUESTION 1): How do the features you made help you distinguish between the 3 classes of traffic light images?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer:**\n",
    "* the first feature detects brightness per row, along the length of the traffic light. The mean brightness can be found and charaterized as red, yellow, or green based on where it is in the image\n",
    "* the second feature multiples the hue and value histograms together to reinforce or question the value filter.\n",
    "* the third feature filters the image by hue color. This can be used to see if any red, yellow, or green pixels have been found, and which is the most dominant in the image. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 4. Classification and Visualizing Error\n",
    "\n",
    "Using all of your features, write a function that takes in an RGB image and, using your extracted features, outputs whether a light is red, green or yellow as a one-hot encoded label. This classification function should be able to classify any image of a traffic light!\n",
    "\n",
    "You are encouraged to write any helper functions or visualization code that you may need, but for testing the accuracy, make sure that this `estimate_label` function returns a one-hot encoded label."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='task8'></a>\n",
    "### (IMPLEMENTATION): Build a complete classifier "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "image:  288\n",
      "guess:  [1, 0, 0]\n",
      "actual:  [1, 0, 0]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3c0866d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This function should take in RGB image input\n",
    "# Analyze that image using your feature creation code and output a one-hot encoded label\n",
    "\n",
    "#boolean flag if hot encoder is double encoded\n",
    "def double_encoded(one_hot_encoded):\n",
    "    if (np.sum(one_hot_encoded) > 1):\n",
    "        return True\n",
    "    else:\n",
    "        return False\n",
    "    \n",
    "# boolean flag if hot encoder has a label\n",
    "def has_label(one_hot_encoded):\n",
    "    if (np.sum(one_hot_encoded) > 0):\n",
    "        return True\n",
    "    else:\n",
    "        return False\n",
    "\n",
    "# hot encode the image based on where the mean of the feature is in relation to the height of the\n",
    "# traffic light.\n",
    "def hot_encode_height(mean):\n",
    "    # bin ranges\n",
    "    # red = 0, 15\n",
    "    # yellow = 10, 23\n",
    "    # green = 20, 32\n",
    "    one_hot_encoded = [0, 0, 0]\n",
    "    \n",
    "    if mean >= 0 and mean < 15:\n",
    "        one_hot_encoded[0] = 1\n",
    "    if mean >= 10 and mean < 23:\n",
    "        one_hot_encoded[1] = 1\n",
    "    if mean >= 20 and mean <= 32:\n",
    "        one_hot_encoded[2] = 1\n",
    "    \n",
    "    return one_hot_encoded\n",
    "\n",
    "def estimate_value(rgb_image):\n",
    "    \n",
    "    feature = feature_value(rgb_image)\n",
    "    \n",
    "    # get mean and bimodal boolean\n",
    "    max_list2 = max_idx_rank(feature)\n",
    "    mean = max_list2[0]\n",
    "    bimodal = is_bimodal(max_list2, feature)\n",
    "    \n",
    "    one_hot_encoded = hot_encode_height(mean)\n",
    "    return one_hot_encoded, bimodal \n",
    "\n",
    "\n",
    "def estimate_hueXvalue(rgb_image):\n",
    "    \n",
    "    feature = feature_valueXHue(rgb_image)\n",
    "    \n",
    "    # get mean and bimodal boolean\n",
    "    max_list2 = max_idx_rank(feature)\n",
    "    mean = max_list2[0]\n",
    "    bimodal = is_bimodal(max_list2, feature)\n",
    "    \n",
    "    one_hot_encoded = hot_encode_height(mean)\n",
    "    return one_hot_encoded, bimodal \n",
    "        \n",
    "    \n",
    "def estimate_color(rgb_image):\n",
    "    feature = feature_rgb(rgb_image)\n",
    "    \n",
    "    one_hot_encoded = [0, 0, 0]\n",
    "    # sum channels representing each light color\n",
    "    red_sum = np.sum(feature[:,:,0])\n",
    "    green_sum = np.sum(feature[:,:,1])\n",
    "    yellow_sum = np.sum(feature[:,:,2])\n",
    "    \n",
    "    # one hot encode the color who has the greatest sum\n",
    "    if red_sum > (yellow_sum + green_sum):\n",
    "        one_hot_encoded[0] = 1\n",
    "    if yellow_sum > (green_sum + red_sum):\n",
    "        one_hot_encoded[1] = 1\n",
    "    if green_sum > (yellow_sum + red_sum):\n",
    "        one_hot_encoded[2] = 1\n",
    "        \n",
    "    return one_hot_encoded\n",
    "        \n",
    "# rules\n",
    "    # * if all hues detected, or white is predominant. color is most liekly yellow.\n",
    "    # * if a bimodal distribution of value vs traffic light length is found.\n",
    "    #     - compute hue * value.\n",
    "    #     - use hue * value mean retrieved from this plot.\n",
    "\n",
    "def estimate_label(rgb_image, print_stats = False):\n",
    "    \n",
    "    ## TODO: Extract feature(s) from the RGB image and use those features to\n",
    "    ## classify the image and output a one-hot encoded label\n",
    "    predicted_label = [0,0,0]\n",
    "    predicted_label_height = []\n",
    "    predicted_label_color = []\n",
    "    bimodal = False\n",
    "    # copy picture\n",
    "    image = np.copy(rgb_image)\n",
    "    \n",
    "    # get hot encode for value vs height\n",
    "    predicted_label_Value, bimodal_value = estimate_value(image)\n",
    "    \n",
    "    if print_stats:\n",
    "        print('height: {}, bimodal: {}'.format(predicted_label_Value, bimodal_value))\n",
    "    \n",
    "    # if bimodal compute hue * value and get the hot encode vs height\n",
    "    if bimodal == True:\n",
    "        predicted_label_height, bimodal = estimate_hueXvalue(image)\n",
    "    else:\n",
    "        predicted_label_height = predicted_label_Value\n",
    "        bimodal = bimodal_value\n",
    "    \n",
    "    # get hot encode for color\n",
    "    predicted_label_color = estimate_color(image)\n",
    "    \n",
    "    if print_stats:\n",
    "        print('height: {}, bimodal: {}'.format(predicted_label_height, bimodal))\n",
    "        print('color: {}'.format(predicted_label_color))\n",
    "    \n",
    "    #union of two hot_encoded labels\n",
    "    union = matrix_multiplication(predicted_label_height, predicted_label_color)\n",
    "    \n",
    "    #if height label and color label match return encoder\n",
    "    if predicted_label_height == predicted_label_color:\n",
    "        predicted_label = predicted_label_height\n",
    "        # check to make sure two colors arn't labeled\n",
    "        if not double_encoded(predicted_label_height):\n",
    "            return predicted_label\n",
    "        \n",
    "    # if height label and color label do not match\n",
    "    #if double encoded\n",
    "    if double_encoded(predicted_label_height):\n",
    "        #use color if not double encoded\n",
    "        if not double_encoded(predicted_label_color) and has_label(predicted_label_color):\n",
    "            # check to make sure estimates do not conflict, and contain a union\n",
    "            if np.sum(union) > 0:\n",
    "                predicted_label = predicted_label_color\n",
    "            else:\n",
    "                #if double encoded on green or red choose red\n",
    "                if predicted_label_height[0]*predicted_label_height[1] > 0:\n",
    "                    predicted_label = [1, 0, 0]\n",
    "            return predicted_label\n",
    "        #if color is double encoded\n",
    "        else:\n",
    "            # if they are encoded the same choose red if not use the union\n",
    "            if np.sum(union) == 1:\n",
    "                predicted_label = union\n",
    "            else:\n",
    "                predicted_label = [1, 0, 0]\n",
    "            return predicted_label\n",
    "    # if height label is not double labelled and conflicts with color\n",
    "    # choose color if available\n",
    "    else:        \n",
    "        if has_label(predicted_label_color) and predicted_label_color != [0, 1, 0]:\n",
    "            predicted_label = predicted_label_color\n",
    "        # if no color got off of value feature\n",
    "        # if bimodal \n",
    "        else:\n",
    "            if bimodal_value:\n",
    "                predicted_label = predicted_label_height\n",
    "            else:\n",
    "                predicted_label = predicted_label_Value\n",
    "    \n",
    "    return predicted_label   \n",
    "\n",
    "image, label, image_num = getImage(color='red', Random=True, showImg=False)\n",
    "#image_num = random.randint(0, total_num_of_images-1)\n",
    "#image_num = 757\n",
    "test_im = STANDARDIZED_LIST[image_num][0]\n",
    "test_label = STANDARDIZED_LIST[image_num][1]\n",
    "\n",
    "encoder = estimate_label(test_im)\n",
    "print('image: ', image_num)\n",
    "print('guess: ', encoder)\n",
    "print('actual: ', test_label)\n",
    "plt.imshow(test_im)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing the classifier\n",
    "\n",
    "Here is where we test your classification algorithm using our test set of data that we set aside at the beginning of the notebook! This project will be complete once you've pogrammed a \"good\" classifier.\n",
    "\n",
    "A \"good\" classifier in this case should meet the following criteria (and once it does, feel free to submit your project):\n",
    "1. Get above 90% classification accuracy.\n",
    "2. Never classify a red light as a green light. \n",
    "\n",
    "### Test dataset\n",
    "\n",
    "Below, we load in the test dataset, standardize it using the `standardize` function you defined above, and then **shuffle** it; this ensures that order will not play a role in testing accuracy.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using the load_dataset function in helpers.py\n",
    "# Load test data\n",
    "TEST_IMAGE_LIST = helpers.load_dataset(IMAGE_DIR_TEST)\n",
    "\n",
    "# Standardize the test data\n",
    "STANDARDIZED_TEST_LIST = standardize(TEST_IMAGE_LIST)\n",
    "\n",
    "# Shuffle the standardized test data\n",
    "random.shuffle(STANDARDIZED_TEST_LIST)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Determine the Accuracy\n",
    "\n",
    "Compare the output of your classification algorithm (a.k.a. your \"model\") with the true labels and determine the accuracy.\n",
    "\n",
    "This code stores all the misclassified images, their predicted labels, and their true labels, in a list called `MISCLASSIFIED`. This code is used for testing and *should not be changed*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9393939393939394\n",
      "Number of misclassified images = 18 out of 297\n"
     ]
    }
   ],
   "source": [
    "# Constructs a list of misclassified images given a list of test images and their labels\n",
    "# This will throw an AssertionError if labels are not standardized (one-hot encoded)\n",
    "\n",
    "def get_misclassified_images(test_images):\n",
    "    # Track misclassified images by placing them into a list\n",
    "    misclassified_images_labels = []\n",
    "\n",
    "    # Iterate through all the test images\n",
    "    # Classify each image and compare to the true label\n",
    "    for image in test_images:\n",
    "\n",
    "        # Get true data\n",
    "        im = image[0]\n",
    "        true_label = image[1]\n",
    "        assert(len(true_label) == 3), \"The true_label is not the expected length (3).\"\n",
    "\n",
    "        # Get predicted label from your classifier\n",
    "        predicted_label = estimate_label(im)\n",
    "        assert(len(predicted_label) == 3), \"The predicted_label is not the expected length (3).\"\n",
    "\n",
    "        # Compare true and predicted labels \n",
    "        if(predicted_label != true_label):\n",
    "            # If these labels are not equal, the image has been misclassified\n",
    "            misclassified_images_labels.append((im, predicted_label, true_label))\n",
    "            \n",
    "    # Return the list of misclassified [image, predicted_label, true_label] values\n",
    "    return misclassified_images_labels\n",
    "\n",
    "\n",
    "# Find all misclassified images in a given test set\n",
    "MISCLASSIFIED = get_misclassified_images(STANDARDIZED_TEST_LIST)\n",
    "\n",
    "# Accuracy calculations\n",
    "total = len(STANDARDIZED_TEST_LIST)\n",
    "num_correct = total - len(MISCLASSIFIED)\n",
    "accuracy = num_correct/total\n",
    "\n",
    "print('Accuracy: ' + str(accuracy))\n",
    "print(\"Number of misclassified images = \" + str(len(MISCLASSIFIED)) +' out of '+ str(total))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='task9'></a>\n",
    "### Visualize the misclassified images\n",
    "\n",
    "Visualize some of the images you classified wrong (in the `MISCLASSIFIED` list) and note any qualities that make them difficult to classify. This will help you identify any weaknesses in your classification algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "height: [1, 0, 0], bimodal: True\n",
      "height: [1, 0, 0], bimodal: True\n",
      "color: [0, 0, 0]\n",
      "guess:  [1, 0, 0]\n",
      "actual:  [0, 0, 1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9b3bd81da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize misclassified example(s)\n",
    "## TODO: Display an image in the `MISCLASSIFIED` list \n",
    "## TODO: Print out its predicted label - to see what the image *was* incorrectly classified as\n",
    "rand_misclass_idx = random.randint(0, len(MISCLASSIFIED)-1)\n",
    "rand_misclass_idx = 17\n",
    "test_im = MISCLASSIFIED[rand_misclass_idx][0]\n",
    "test_label = MISCLASSIFIED[rand_misclass_idx][2]\n",
    "encoder = MISCLASSIFIED[rand_misclass_idx][1]\n",
    "\n",
    "encoder = estimate_label(test_im, True)\n",
    "print('guess: ', encoder)\n",
    "print('actual: ', test_label)\n",
    "plt.imshow(test_im)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a id='question2'></a>\n",
    "## (Question 2): After visualizing these misclassifications, what weaknesses do you think your classification algorithm has? Please note at least two."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer:** if light has not been cropped at the top, and there is a siginificant brightness at the top, for the height feature it will distinguish this in the value plot. THe height feature has a significant error probablity, where the hue feature was used as validation. However the hue ranges between the three colors are not  always distint, and cannot be filtered easily. greater precision in the hue filters would be needed to increase the accuracy of this model. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test if you classify any red lights as green\n",
    "\n",
    "**To pass this project, you must not classify any red lights as green!** Classifying red lights as green would cause a car to drive through a red traffic light, so this red-as-green error is very dangerous in the real world. \n",
    "\n",
    "The code below lets you test to see if you've misclassified any red lights as green in the test set. **This test assumes that `MISCLASSIFIED` is a list of tuples with the order: [misclassified_image, predicted_label, true_label].**\n",
    "\n",
    "Note: this is not an all encompassing test, but its a good indicator that, if you pass, you are on the right track! This iterates through your list of misclassified examples and checks to see if any red traffic lights have been mistakenly labelled [0, 1, 0] (green)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "**<span style=\"color: green;\">TEST PASSED</span>**"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Importing the tests\n",
    "import test_functions\n",
    "tests = test_functions.Tests()\n",
    "\n",
    "if(len(MISCLASSIFIED) > 0):\n",
    "    # Test code for one_hot_encode function\n",
    "    tests.test_red_as_green(MISCLASSIFIED)\n",
    "else:\n",
    "    print(\"MISCLASSIFIED may not have been populated with images.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Improve your algorithm!\n",
    "\n",
    "**Submit your project after you have completed all implementations, answered all questions, AND when you've met the two criteria:**\n",
    "1. Greater than 90% accuracy classification\n",
    "2. No red lights classified as green\n",
    "\n",
    "If you did not meet these requirements (which is common on the first attempt!), revisit your algorithm and tweak it to improve light recognition -- this could mean changing the brightness feature, performing some background subtraction, or adding another feature!\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Going Further (Optional Challenges)\n",
    "\n",
    "If you found this challenge easy, I suggest you go above and beyond! Here are a couple **optional** (meaning you do not need to implement these to submit and pass the project) suggestions:\n",
    "* (Optional) Aim for >95% classification accuracy.\n",
    "* (Optional) Some lights are in the shape of arrows; further classify the lights as round or arrow-shaped.\n",
    "* (Optional) Add another feature and aim for as close to 100% accuracy as you can get!"
   ]
  },
  {
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}