3-8-16-Assignment.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "xfNT-mlFwxVM"
      },
      "source": [
        "# Autoencoders for Anomaly Detection"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "ITZuApL56Mny"
      },
      "source": [
        "In this Assignment, you will again train an autoencoder to detect anomalies on the [ECG5000 dataset](http://www.timeseriesclassification.com/description.php?Dataset=ECG5000).\n",
        "\n",
        "This time 10% of our training set is anomalies the labels are not available at training time. This is to reflect a truely unsupervised scenario where we may not have access to labeled data or a expert knowledge. Since the majority of our training data is still normal, we are still able to train a high performing model as long as we don't overfit to our training data.\n",
        "\n",
        "In this assignment you will **first select the size of the encoding layer and try to maximize the AUC metric**. Be careful, if the model is too large it can learn to recreate the abnormal data as well as the normal data, and therefore do a worse job a distinguishing them.\n",
        "\n",
        "Next, you will **pick a error threshold to maxmize accuracy, precision, and recall.** There is often a tradeoff between high precision and high recall, therefore it is up to you to decide what is important to our ECG classification application. Remember that our model predicts `1` when it predicts a normal rhythm.\n",
        "\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "e1_Y75QXJS6h"
      },
      "source": [
        "### Import TensorFlow and other libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "if tf.__version__ != \"2.10.0\":\n",
        "    print(f\"Current TensorFlow version: {tf.__version__}, switching to 2.10.0\")\n",
        "    \n",
        "    # Uninstall current TensorFlow version\n",
        "    !pip uninstall -y tensorflow\n",
        "    \n",
        "    # Install TensorFlow 2.10\n",
        "    !pip install tensorflow==2.10\n",
        "    \n",
        "    # After installation, restart runtime\n",
        "    print(\"TensorFlow 2.10 installed.\")\n",
        "    print(\"Please click on the Runtime > Restart session and run all.\")\n",
        "else:\n",
        "    print(\"TensorFlow 2.10 is already installed.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "YfIk2es3hJEd"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "\n",
        "from sklearn.metrics import accuracy_score, precision_score, recall_score, roc_curve, auc\n",
        "from sklearn.model_selection import train_test_split\n",
        "from tensorflow.keras import layers, losses\n",
        "from tensorflow.keras.models import Model"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "i5estNaur_Mh"
      },
      "source": [
        "### Load ECG data\n",
        "\n",
        "Same as the previous colab"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "y35nsXLPsDNX"
      },
      "source": [
        "The dataset you will use is based on one from [timeseriesclassification.com](http://www.timeseriesclassification.com/description.php?Dataset=ECG5000).\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KmKRDJWgsFYa"
      },
      "outputs": [],
      "source": [
        "# Download the dataset\n",
        "dataframe = pd.read_csv('http://storage.googleapis.com/download.tensorflow.org/data/ecg.csv', header=None)\n",
        "raw_data = dataframe.values\n",
        "dataframe.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "UmuCPVYKsKKx"
      },
      "outputs": [],
      "source": [
        "# The last element contains the labels\n",
        "labels = raw_data[:, -1]\n",
        "\n",
        "# The other data points are the electrocadriogram data\n",
        "data = raw_data[:, 0:-1]\n",
        "\n",
        "train_data, test_data, train_labels, test_labels = train_test_split(\n",
        "    data, labels, test_size=0.2, random_state=21\n",
        ")"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "byK2vP7hsMbz"
      },
      "source": [
        "Normalize the data to `[0,1]` to improve training accuracy.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "tgMZVWRKsPx6"
      },
      "outputs": [],
      "source": [
        "min_val = tf.reduce_min(train_data)\n",
        "max_val = tf.reduce_max(train_data)\n",
        "\n",
        "train_data = (train_data - min_val) / (max_val - min_val)\n",
        "test_data = (test_data - min_val) / (max_val - min_val)\n",
        "\n",
        "train_data = tf.cast(train_data, tf.float32)\n",
        "test_data = tf.cast(test_data, tf.float32)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "BdSYr2IPsTiz"
      },
      "source": [
        "This time we are mixing in some of the anomaly data into the training set. \n",
        "\n",
        "10% of the training data will contain anomalies. Since the majority of the training data is still normal data, we can still train a high performing model as long as we don't overfit to the training data."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VvK4NRe8sVhE",
        "outputId": "529b2266-5936-47d7-e108-6e543fb6af63"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(2621, 140)"
            ]
          },
          "execution_count": 5,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_labels = train_labels.astype(bool)\n",
        "test_labels = test_labels.astype(bool)\n",
        "\n",
        "normal_train_data = train_data[train_labels]\n",
        "normal_test_data = test_data[test_labels]\n",
        "\n",
        "anomalous_train_data = train_data[~train_labels]\n",
        "anomalous_test_data = test_data[~test_labels]\n",
        "\n",
        "portion_of_anomaly_in_training = 0.1 #10% of training data will be anomalies\n",
        "end_size = int(len(normal_train_data)/(10-portion_of_anomaly_in_training*10))\n",
        "combined_train_data = np.append(normal_train_data, anomalous_test_data[:end_size], axis=0)\n",
        "combined_train_data.shape"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "wVcTBDo-CqFS"
      },
      "source": [
        "Plot a normal ECG. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "ZTlMIrpmseYe",
        "outputId": "5aa32780-5226-4589-80df-9a69fa6a7dcd"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3ib1dn48e8tyXsmceIkziSDbMgggwRImAkjzJZVRsvsWwqFlkKhLy8/eNvSUlp2gbL6ssIoMyQNEGIggewEsmPHznISO8N7yZLP7w9JjpJ4yLZkSY/vz3X5ws+QdOfBunV0n/OcI8YYlFJKRT9buANQSikVHJrQlVLKIjShK6WURWhCV0opi9CErpRSFqEJXSmlLEITulLtICKvisj/hjsOpUATugoTEckWkWIRiQvgvBoR6eu370wR2R7yINtJRK4XEbeIVBz109vvnKtEZKV3/14RmS8i0/yODxGROSKyX0TKRCRHRJ4SkT7h+VepSKYJXXU4ERkAnAIYYHYAD6kE/jtIr20PxvO0wnfGmOSjfvZ4Y7kLeBz4I5AJ9AOeBS70Hh8MLAP2AGONManAVGAbMO3Yl1KdnSZ0FQ7XAkuBV4HrAjj/SeBKERnU2EERGe5tyZeIyAYRme137FUR+YeIzBORSmCGiGwXkbtF5AcRqRSRl0Qk09s6LheRL0Ski99zvCsi+0SkVES+FpGR7frXe54zDXgI+IUx5n1jTKUxps4Y84kx5m7vaQ8CS4wxdxljdgMYY4qMMY8bY+a0NwZlPZrQVThcC7zh/TlHRDJbOL8A+Cfw/44+ICIxwCfAZ0AP4JfAGyJyvN9pVwF/AFKAxd59lwJnAUOBC4D5wH1Adzzvi9v9Hj8fGOJ9/tXeuNtrChAPfNDMOWcC/w7Ca6lOQhO66lDe+nB/4B1jzCo85YOrAnjon4ALGmkdTwaSgUeMMU5jzJfAXOBKv3M+MsYsMcbUG2NqvPueMsYUGmMKgG+AZcaYNd7jHwBjfQ82xrxsjCk3xtTiaTWf4G1hB2Ky95uD72ebd3834IAxxtXMYzOAfb4NEbnN+xwVIvLPAF9fdSKa0FVHuw74zBhzwLv9JgGUXYwx+4Gn8ZQp/PUGdhlj6v327QCy/LZ3NfKUhX6/VzeynQyemruIPCIi20SkDNjuPSejpZi9lhpj0v1+fGWjg0CGiDiaeexBoJdvwxjztDEmHU/dPSbA11ediCZ01WFEJAH4MXCatya9D7gTT4v3hACe4lFgBjDeb98eoK+I+P8t98NTpvFpz5SiV+HppDwTSAMGePdLO54T4DugFriomXMWApe083VUJ6IJXXWkiwA3MAI40fszHE/J49qWHmyMKQEeA37rt3sZUAX8VkRiRGQ6npp4sDoNU/Ak3oNAIp4RKe1mjCkFHgCeEZGLRCTRG/8sEfmL97QHgVNE5G8ikgUgIhl4rplSx9CErjrSdcArxpidxph9vh88pZSrWyg/+DyB50MBAGOME08CnwUcwDPs71pjzOYgxfx/eEo4BcBGPKNzWmNKI+PQT/LG/hhwF/B7YD+e0tBtwIfe41uBSUAf4HsRKQeW4PlWEpRhnMpaRBe4UEopa9AWulJKWYQmdKWUsghN6EopZRGa0JVSyiICGVUQEhkZGWbAgAFtemxlZSVJSUnBDSiENN7QiqZ4oylW0HhDrS3xrlq16oAxpnujB40xYfkZP368aatFixa1+bHhoPGGVjTFG02xGqPxhlpb4gVWmibyqpZclFLKIjShK6WURWhCV0opi9CErpRSFhFQQheRmSKyRURyReTeJs75sYhs9K4Y82Zww1RKKdWSFoctetdgfAbP6i67gRUi8rExZqPfOUOA3wFTjTHFItIjVAErpZRqXCAt9IlArjEmz3hmtpuDdxFbPzcBzxhjisGz7mFww1RKKdWSFmdbFJHLgJnGmBu929cAk4wxt/md8yGwFc+K5HbgQWPMfxp5rpuBmwEyMzPHz5nTtimrKyoqSE5ObtNjw0HjDa1oijeaYgWNNxDGGL7e7cJlICNBGNrFToIjsPVP2hLvjBkzVhljJjR2LFh3ijrwLKI7Hc/czV+LyGjjWZCggTHmBeAFgAkTJpjp06e36cWys7Np62PDQeMNrWiKN5piBY03EAs3FfLKgpUN2xed2JvHrxjbzCMOC3a8gZRcCoC+ftt9OHJ5L/DU1j82xtQZY/LxtNaHBCdEpZSKTHXuev4wbxPHZSTx7b2nc8awHizLPxS2eAJJ6CuAISIyUERigSuAj48650M8rXPfEllDgbwgxqmUUhHnzWU7ydtfyX3nDqd3egKnDMlgb2kNe0qqwxJPiwndGOPCsyzWAmAT8I4xZoOIPCQis72nLQAOishGYBFwtzHmYKiCVkqpcCupcvL4F1s5eVA3zhjuGdg3rn8XAFbvLA5LTAHV0I0x84B5R+17wO93g2dtxLuCGp1SSkWg+nrDr95eS0Wti9+fNwIRTyfo8F6pxMfYWLWjmPPH9O7wuPROUaWUaqXHF+aQvWU//3PBSEb0Tm3YH2O3MaZPOqt3hKeFrgldKaVaYXn+IZ5cmMOPxvfh6kn9jjk+vn8XNuwpo6bO3eGxaUJXSqlW8NXH//uCw6UWf+P7dcFVb/hhd2lHh6YJXSmlWqOsug6HTUiJa7wLcmy/dCA8HaNRl9C/3rqfl9fX4q5v/g5XpZQKhbKaOlITYhptnQN0S45jYEYSq8JQR4+6hL61sJyvd7uodLrCHYpSqhMqq3aRGt/8AMERvVPJKSzvoIgOi7qEnuK9kBU1mtCVUh3P10JvTtfEWEqr6zooosOiLqEnx3kuZEWtJnSlVMcrq64jNb75hJ6WEENZjYuWJj8MtuhL6N4Werm20JVSYVBW4yI1ofmSS1pCDO560+ENz+hL6HF2QFvoSqnwCLSFDnR42SUKE7rnQlVqQldKhUFZTV1Dwm5Kqib0wCRrp6hSKkxqXW5q6upb7BQNVws9WAtcdJhk72D+cm2hKxXxjDG8tnQHhyqd9OuayNTBGWSmxoc7rDbz9d21NGzRl9DLNKE3z5fQtYWuVOT762dbeGbRtobtaYMzeP3GSWGMqH18CbrFFnqillwCYrcJcXaoqO34MZ5KqcC9vnQHzyzaxpUT+7L54ZlcPqEvq3cW43LXhzu0NitraKFHZskl6hI6QIJDdJSLUhEst6icBz5azxnDevDwhaOIj7EzZVA3qpxuthZWhDu8NjvcQm++uJEUa8duE03ogYi36zh0pSLZ2yt2YRPhz5eNwWH3pJlx/Tyr+azZFZ65woOhrMab0FtooYsIaQkxlFRpQm9RgkN02KJSEcrpquf91QWcOTyTjOS4hv19uybQLSmW1TtKwhhd+5RVe0suLdTQwVN20RZ6AOIdemORUpHqy81FHKx0cvlJfY/YLyKM7delU7TQwZP0NaEHIMEhWnJRKkK9u3IXmalxnDIk45hjY/ulk7e/kpIqZxgia7+y6jpi7EJ8TMupMy0hpsOHLUZtQtcWulKRp7CshkVbirh0XJ+G2rm/w3X06Cy7lNV4bvtvai50f1pyCZCWXJSKTAs3FVFv4JJxWY0eH9MnDZvAmp1RmtCrXQHVzwHSEhya0AOR4BAqwjA1pVKqeesKSkmNdzCoe3Kjx5PiHBzfM5U1YVieLRg8LfTA7scMxxS6UZnQ4x3gqjfUuqL3BgUVuaqcLp5amMPe0uqGffX1JqpviOkoG/aUMiorrdmSxNh+6azdVRKVDbKy6pYXt/BJT4jt8Cl0ozKhJzg8fyxadlGh8NAnG3ns861c89Jyiiud5BaVM+OxbO6YszbcoUU0p6uezXvLGZ2V1ux5o3qnUV7jYndxdbPnRaKyGldAI1wgPHeLRndC15EuKsj+s34vc1bsYubInuw8VMVVLy7jkme/ZdehKj5dt5etYVgnMlrkFJXjdNczsoWEPrJ3KuBpzTfGGMOrS/LJ3lIU9Bjby9NCD6zkEo4pdKM0oXv+qy10FUwFJdXc+/46xvRJ48krx/LUlWPZsq+MHqnxfPSLacTH2Pjn13nhDjNibSgoA2CUN2E35fieKdhtwoY9ZY0e/8dX23jwk408uTAn6DG2l2+USyDC0UKPutkWAeLtnha6jkVXwbLzYBVXvbgUl9vw+OUnEuuwcc7Innx256n0SksgKc7Bjyf05a3lO/nNOceTvaWIzzYUcsqQDM4b05vuKZ47IitrXTz/1TYqnW6mHNeNKYO6kRQXlW+zVltXUEpynIMB3ZKaPS8+xs6g7kmNJvS3lu/kL//ZQkq8g/UFZdS63MQ57KEKuVUCnQvdJxxT6EblX5q20FWwGGNYnn+IO+aspcbl5o0bJ3Gc3wiNwT1SGn6/cdpxvL50Bxc8tZii8loykuNYuLmIh+ZuZOrgDE4d0p1/fbedgpJqYuw2Xlqcz4heqcz95TRstpbHLUe79XtKGdE7NaB/68jeaXy37eAR+7YfqOT3H67ntKHduWx8H3751ho27ClrGLseboHOhe4Tjil0Ayq5iMhMEdkiIrkicm8jx68Xkf0istb7c2PwQz0svqFTVKfQVW33wZrdzPhrNpe/sBS3Mcy5eTIn9E1v8vx+3RK54ITelNXU8eAFI1h+3xl8fuep/Nf0wWw/WMkf5m0i1m7j3Vum8MP/nM395w5n494yvsk90OjzHaio5aevLOcPn27EGeUjtlzuejbtLWNU7+br5z4je6eyr6yGgxW1Dfue/DKHGLvw6I/GMGlgVwBW74ic4Y2BzoXuE5ElFxGxA88AZwG7gRUi8rExZuNRp75tjLktBDEe4/AoF3dHvJyyoP+s38dd73zP6Kw0Hr1sDLNG92pYPKU5f7lsDA/NHtXQ+hqSmcJvzjmeX589lLwDlWSlJxAf4ykRXHtyf57/Oo9Xl+Rz2tDuRzxPblE517+ygqKyWhZt2c/KHcX87ccnMjCj+XJFpNq2v5KaunpGZTVfP/cZ0dAxWsapQ7uTW1TBh2sKuPGU4+iR4lnRKCs9gdURNF490LnQfcIxhW4gLfSJQK4xJs8Y4wTmABeGNqzmNZRctIau2mDtrhJ+9fYaTuybzju3TOFHE/oGlMwB4hz2hmTuT0QY1D25IZn7zr16Uj8WbdlP/oHKhv25ReVc8uy31NTV8+6tU3j26nHkFFYw46/ZTH3kSx74aD3VzuhqrKwv8IxYaWnIos+IXocTOsCTC3OIj7Fzy6nHNZwzrn+XiJqZMdC50H18U+hGVAsdyAJ2+W3vBhpbQ+pSETkV2ArcaYzZdfQJInIzcDNAZmYm2dnZrQ4YoLaqErsIG7ZuI5tjXibiVFRUtPnfGg5WjrfWZbjnm2pSHPDTQU6WLvkmpLENdNdjF/jju4u5engcRSUV3PPcN1BvuGdcDMXb1pIIPDg5htWFNrYWO3ntux18s3EXd4yLIz0uvAPRAr22731fQ7wddm1cScGmwPoLusULi9bmULc/n0++r+XcgTGsW/ldw/FUZx37ypz8e/6XdEsI7DqE8m93+V5PA3LzurWU5wcWT4ypI2d7AdnZBxs9Hux4g9Up+gnwljGmVkRuAf4FnH70ScaYF4AXACZMmGCmT5/ephfLzs4mJcFJt8zeTJ8+qu1Rd5Ds7Gza+m8NByvH+2x2LiW1W3jv1ilMGNA1tIF5LSpZw/x1+4hNTSVndzWHag1v3XTs61/m/e9nG/Zx+5w1PLoG5v7yZLokxXZInI0J5NruOlTF8s+yuW7KQE6fMSLg5x6/cyVL8w6yZr+T4b1S+eO1kxvqzgBdd5fwxqYlxGUNY/qY3kGLt632LNsJ36/jzFNPDnih654blhCXEMP06RMbPR7seAP5mCkA/Cc27uPd18AYc9AY4+vdeBEYH5zwmpYc56BcR7moZhhj+GhtATf930pyi8opra7j+a/yOH1Yjw5L5gD3nzecyyb0YVn+QbYW1/PwhaOaff2zR/bkjRsnsae0mpcW5zfsr4vQqQf++U0eNoGbTh3YqseN7J1GWY2L0VlpvHXzkckcYHivVOJjbBFTdmnNXOg+kVhyWQEMEZGBeBL5FcBV/ieISC9jzF7v5mxgU1CjbERynENr6KpJOYXl3P/BepZvP4RNYFneQaYM6kZpdR2/Pntoh8bSIyWeP148mocvHMUHCxZx2cR+LT5mfP+unDuqF69+u50bTxlIeY2Ly577lgvG9Ob35wfeCg61/eW1vL1iFxePzaJXWkKrHnvlpL6IwA3TBjY6Vj/GbmNMVjqrIqRjtDVzofukJcSw61BVCKM6UouRGWNcwG3AAjyJ+h1jzAYReUhEZntPu11ENojI98DtwPWhCtgnOc6h49DVMYwxvLFsB+c/tZjc/RU8cslosn8zg8zUeBZsKOT8Mb0YGeDQumCz24SMAGvBAL88YzAVtS6eWZTLLa+torCslpeW5LNy+6FmH2eM4ZlFuWza2/idmMGy42AlD83diNNdzy2nDWr143ukxHP7GUOavfFq/IAubCgojYhO4tLqOlICnAvdp6On0A2ohm6MmQfMO2rfA36//w74XXBDa15yvIPiyuhc9UQF37K8g3y2sZDvth1k494yThmSwWM/PqFhCNx7Pz+Zlxfnc2UAreNIMaxnKrNG9eSf3+QjAs9cNY4/ztvEve+v49PbpzV5B+V32w7y6IItfPrDXj755TTsQb6pyV1vuPX1VXy+sRAR+NnUgU1Ol9teEwd05R/Z21izq5iTBx27AlJH2lVcTe/0wGrnPr6SizGmVR8EbRWVc7mA1tCVR02dmzc21XL5C0t5bekOkuMdPHThSP7104kNyRw8b6w7zxpKz7TWvSHD7fYzhpAQY+e35wzjvDG9+N+LR5FbVMHfP296npPnvs4j1m5j494y/r1qd9Bjemv5Tj7fWMitpw3i23tP579DWAIa168LIrBye/jLLrmF5Qzxu3M4EF0SPVPolnVQeTgqb/0HSIkPfQ3dGMOqHcW8v6aAS8dlMb7/kR1ZlbUuPlxbwPqCUnYXV/PHi0fTt2tiSGNSh9XUufnRc9+xrsDFT6cO4J6Zw44YB24Fw3ulsuaBsxr+XTOO78HlE/ry3FfbcNiEX5899IiW36a9ZXy9dT+/OXsoCzcX8ehnWzhvTK82zSezakcx2bvqmOaub1hO7mBFLY8u2MKU47pxz8zjQ97qTEuM4fjMFFa0UGYKtYpaF3tKaxjco3XfRHxz/ByoqD2m0zcUojahh7qGXlReww2vrmSd94aJb3MPsODOUxu+5tbUufnpKytYvv0QaQkxVDvd/P3zrfzt8hNDFpM60sdr97CuoJSbx8Rx3wUjwx1OyBz9IfXHS0Zjs8HTi3LZXVzFVZP6M75/F+w24YWv80iMtXPN5AGcPDiDS579lvs+WMeDF4xsGP7Y0tf/nMJy/vfTTXy1db9n+5XlPHnFWJLiHPxx3mYqa108dOHIDikhAJw0oCvvr96Ny++DpaNtK6oAaH1CT/Yk9P3ltQzqnowxhvwDlfT2u6M4mKI4ocdQ5XTjrjdBrxEaY7j/g/VsLSznDxePoltSHLe+vopXlmzn1tMG4XLXc9ubq1mx4xBPXHEis0/ozSPzN/PCN3n8fPoghmS27muZOswYw4INhZw0oAvdvG+Gkion5TWuI779GGN4eUk+w3qmMKVX+DvMOpLdJvzx4tF0TYrlha/z+HDtHhJi7KTEOzzzw0wdSFpiDOP6deEXMwbxj+xtfLmpiIkDu7JxbxmuesPz14xvdNKr4kon17y0nBqXm3tnDaNwZx5vbinmtEezqa7zvN9uOfW4Dv0bnzCgC68t3cHmfeWMCvBO1GDLbWNCz0g5nNABDlY6Of2xr3jwghFcP7V1wzwDEb0J3TvjWUWtK2hfZerc9cTYbXy0dg+fbyzk/nOHc/Wk/gCcObwHTy3MoW+XRF79Np8V24t5+MKRXHiiZzHcW04bxOtLd/D4Fzk8c/W4oMTTGX2wpoC73vmemSN78tw14zHG8LNXV7Bxbxmv3TCJk7zjt7/bdpDN+8r5y6VjkMptYY6644kId58zjFtPG8Q3OQdYub2YKqfnG+utfiNO7j5nGBedmMVjn21la1E5Jw3oytpdJVz30nL+dcPEI5K6MYb7PljHwcpaPvivqYzKSiM7exeXnj6Rl5fk0zstgVFZqZw1omeH/lsneifqWp5/KGwJPaeoghi70L+VJVVfC/2AdxKyfaU1APRs5RDPQEVvQo/zfF2pDFJCf/6rbTy6YAtj+6WztbCCcf3S+dm0w5+gvz9vBGf9/St+8eZqeqTE8cglo7nCb8RE16RYbpg2kCe/zOXnBaVh+8OLZrsOVfHARxtIiLHznw372LS3jD0l1azeWUJynIOfvbKCN2+azOg+aby8JJ+uSbHMPrE3S5d0voTukxIfw7mje3Hu6F5NnjMkM4Xnrjl8r9/e0mqueGEp1720nHl3nNLwzeedlbuYv34fv5s17Ii/31FZafztx+ErJfZKSyArPYGVOw4d8Z7sSLlFFQzMSGp1ySctIQaHTRpa6IcTemg656N4lIsniQerjr5hTxkJMXZqXfXeKTxPOKKUMyAjiSevGMufLx3NN/fMOCKZ+9xwynF0TYrld++vi9i7+iJNTmE5D3y0nsc+28Jtb65GgPd+PoWUOAePf7GVv362lf7dEpl/xymkJsRw0bNLmPKnhSzcXMRPJvWzXCdoR+iVlsDrN0yiwuni3ZWeuZBq6tz8af5mJg3syk2nHNfCM3S8iQO7sjy/OGwLS+cWtX6EC4DNJnRLjm1ooe8t8yT0XiFK6FHbQk/yttCDtWpRYVkNw3ul8s6tU5o8Z1YzrSDwfBr/4aJR/PyN1Tz9ZS53ntWxdyRGmx0HK7nqxWWUVtfhctdjE+GxH5/AyN5p/HTawIYlyP5++Qn07ZrI27dM5s1lOyksq6XG5ea6kweE9x8Qxfp2TeTkQd346Ps93HnWUD7bWEhJVR23nT44IhfjGNsvnQ/WFLCvrKbVd6S2V02dm52HqpjtLa+2VveUuIYWemFpjecGM28pJtiiNqEneFtmtXXB6RDbX17bMEdze8wa3YtLxmbx9KJcEmLtuNz1uA64mN7+EC2lqKyGa15ajstdz7zbpzGgWxI1rvqGaWxvmDqQVxbn0zMtntkneN5Ifbok8tuZw8IZtqVceGIWv33vB9buKuGdFbvISk9gaphv3mnKUG8n7NbCig5P6PkHKqk3MKSVHaI+GclxHKjw3AS5t7SGHilxQR/I4RO1JRffV+3aIK30UlhWc8SNKO3xP7NHkpWewCPzN/PXz7by3Pe1DR1WytP5fMvrqzhQUcsrP53I4B4pOOy2I+YkT0uM4c2bJvPPayeE7I+/s5s5qiexDhvPLNrGkm0H+NGEPhHZOofDyTSnsLzDXzunjSNcfLon+7XQy2pCenNb1Cb0OO8EOTVBaKFX1LqodLrJTA3O16C0hBg+u/NUlt13Bq/fMIkaN8xbt++Ic6qdbj5aW0BplXWX0atyunh47saGxQ98Hl2whTU7S/jLZWM4sZkl30b3SWNAlK7gEw1S42M4Y1gPvthUCMCPJvRt4RHh0y05jm5JsWwNQ0LPLarAJrR5NamMlDgOVtZSX2/YW1pNzwCn3m2L6E3ojuC10Iu8HRWBznEciPgYO5mp8Uwd3I3MRGnofHK563nxmzxO+cuX3DFnLbe8vhKXRTtQ316xi5cW53PpP77lwzUFFFc6eWfFLl74Oo+fTO7H+QHOca1CxzfsdtrgDLLSO7aU0VpDMpPZWljR4a+bW1ROv66Jbe6A754cR53bUFpdR2FZbUhb6FFbQ49zeD6Lal3tb6EXlnm+DvVICX5HhYgwNcvB+zmH2HGwkue/zuPNZTuZNjiDcf278OTCHP40f3NI58MIB3e958af0VlpJMTa+dXbaxuOjcpK5ffnWevfG61mDOvOmcN7ROTIlqMNzUzh/dUFHTbRFXiGeH6Tc4BTj1oTtjV8NxflH6ykotYV0hZ61CZ036dlTV0QWujlnhZ6jxBd6GlZDj7IrePm/1vFlsJyfj59EPd4O/fKqut4aXE+I3qlcun4PiF5/XD4fOM+dh2q5r6rh3PmiEz+9e123PWGEb1TmdC/qw43jBBxDjsvXndSuMMIyJDMlIY5VTri20R9veHX73yPu97w23OOb/Pz+G4u8pUetYXeiGC20Iu8LfRg1dCP1jXexilDuvP11v3MGtWTu88+/Mdx/3nD2VpYzt3vfU+sw8YFJ0RPGWLBhn3sOFhJekIs3VPi6Ns1gT5dPF9NX/wmn75dEzh7ZE/sNuHGKGgBqsg21K9jtCMS+stL8vl220H+fOlo+ndre19O9xTPHDrrdnsSeihH6UR/Qg9CC72wrIaEGHvAK7+3xd1nH0+/rgncf+6II0YSxNhtvHjdBK5/ZQW/enstDps0jHevdblZtaOYKcd167CvmIGqqXPzyzfX4Gyk/u8ZplXL/1wwQkeoqKDxDV3MKaxg+vE9Qvpam/eV8Zf/bOHsEZn8uJ2dxd2TPS1y30R/WnJphMNuw26ToHSKFpbXkpkaF9KkObpPGqP7jG70WGKsg1euP4lrXlrGb979nrH9utAzLZ4/zdvMq99u58krxzI7wlruq3cU43TX8+zV4xjTJ43Cslp2Hary/BRX4XTVt/uNoJS/LkmxZCTHhXykS02dm1/NWUtqQgx/umR0u/NCaoKDWLutYfhjjxBVAiCKEzpAvMMWlGGLhWU1IaufByopzsHfLz+Rs//+NQ/P3cjNpx7Hv77bjk3gsc+2MHOkZ8xwpPgu7yB2m3DKkAxS4mPo0yWR8f2Pnb1PqWAampnM1qLgjnRx1xteWZLPJz/s5eRB3ThU4WTzvnJeuf6khhk/20NEyEiOZU9pDV2TYkPafxQ5GaIN4rxzr7TX/vLaoA5ZbKv+3ZL4xYzBfLpuL7e8toruyXE8fsVYdhys4u0VO8Md3hGW5h1kVFYaKa1YAV2p9hqamUJuYXnQ5nTJP1DJj577lv/9dBNVtS5e+DqPt1fu4ieT+zFjWPDKOr6RLqEst0CUt9DjHLZ2d4oaYygsq+H0IP7Pa49bTjuOD9cUkHegkmevHsesUfzGsVoAABkmSURBVD15fekOnliYy+wTskhLDH8CrXa6WburJGwz36nOa0hmMpVONwUl1fTpcuxUtsaYhomwWpov5fONhdz19lpsNuHxy0/kwhN7c6jSyfL8Q0FN5nB4pEuoJuXyieqEHh9jb/ewxYpaF1VBvEu0veIcdp67ZjzL8g8xa1RPRIR7Zw3jsn98y8mPLOTicVncccbQhqWtwmHVjmLq3IbJx3ULWwyqcxrV2zOt77K8Q/QZf2RCf2ZRLk98UYVzwRckxzn48jenHTOdhzGG9QVlzFmxkzeW7WR0Vhr/+Mm4hg+HbslxLU7C1xa+D5fMECf06C65BKGFXlTuG7IY/pKLz9DMFK6Z3L+hM2Zcvy58fNs0Zo3uxdsrdvHw3I1hjW+pt37uW2xCqY4yOiuNrPQE5v6w54j9hyqdPLkwh4GpNu6ZOYzqOjfPLjpynvzCshouevZbLnh6Me+u3M1PJvfj3VunNNrSDzZfA6yXllya5kno7WuhF3pv+w9nizcQo7LS+OuPTiDWYeP91bupqHWFdJhlc5bmHWR0VlrYXl91XjabcN6YXry8OJ+SKifpiZ4x3nNW7KTWVc+1IxK4evogdhys5M1lO7np1OPISk9g2/4Krn1pOSVVTh6+aBSzx/Tu0PJlRrInTm2hNyPOYW/3OPTDNxVFTgu9OZeMzaKmrp756/aG5fUPVTr5fneJlltU2FwwpjeuesN/1nsmvHO563ntux1MHdyNrBRPSvvlGUMAeGT+Zv7++VYufmYJtS43b98yhWsm9+/wvihfp2ioa+jRndBjbNS0u+QS/Im5Qml8/y7065rIB2sKOvy1Xe56bn9rDYJE3Lh41XmMykplQLdEPvGWXT7bWMje0hquP/lwJ31WegJXTerHJ9/v4YmFOZw0oCv//vnJYVsacuqgDK44qW+jC3MHU1R/Z45z2DnonTi+rQrLakmMDe1dosEkIlw0Nounvsxhb2l1yCf7Lyqr4YtNRQzNTGb++n0szj3Any8dHZTFQJRqCxHh/DG9eTY7l5cW5/Pad9vp2zWB04f14JuiTQ3n3XnWUHqmxXP2iEyO6962ucyDpUtSLI9cOibkrxP1LfT2dooWltVETevc5+KxWRgD76zYHfLXevCTDdz3wToue+47Xlqcz08m9+Pyk45dT1WpjnTBCb2pN/Dw3I0Y4KHZo46ZZiItIYZbTxsU9mTekaKjWdqEOIet3cMW95fXRnyH6NEGZiRx8qBu/P2LrWRvLeL204cEfdwsQN7+Cuav38f1Jw9g2uAM9pXV6O38KiIc3zOF//vZRDJT4xmamRxxcx2FS1S30OODcKdocZWTrt6e8mjyz2sn8OAFIzhU6eSGf61g5fZDAGzbX8EVL3zH3z7fSkFJdbte4/mv8oi127jt9MGcOSKTn0zuH1HTD6jO7dSh3Tm+Z4omcz9R/e4Mxjj04qo6uiSF/+7L1kqKc3D91IHM/eU0srokcMecteQWeYZm/bC7lKe+zGHan7/kT/M3UV8f+G3Sbyzbwfz8OtYXlPL+mt1cflLfkK1QrpQKroASuojMFJEtIpIrIvc2c96lImJEZELwQmxanKN9LXRjDCVVTtISoq+F7pMSH8MTV4xlX1kNs574muIqJ3Nunsw3v53Bj8f35fmv8vivN1ZT7Wz5gy+nsJzff7iet7c4Of+pxdQbomIlG6WUR4sJXUTswDPALGAEcKWIHLN+mIikAHcAy4IdZFPiY2w4XfWtaoH6q3S6qXMbukTA/CjtMa5fF+4+53hExDudbTp9uiTyyKWj+e/zR7Bg4z5u/L8V1LWwdunTi3JJiLFz/6R4bj1tEL8/bzh9u4b+LjqlVHAE0ik6Ecg1xuQBiMgc4ELg6PvPHwb+DNwd1Aib4Vso2umuJ97W+ikpiys9Qx67RGEN/Wi3njaI608ecMTUnCLCDdMGkhLv4Lfv/cAfPt3Eg7NHNvr4bfsr+OT7Pdx06nEMSSjkpunDOip0pVSQBJLQs4Bdftu7gUn+J4jIOKCvMeZTEWkyoYvIzcDNAJmZmWRnZ7c6YICKigqys7PZtb0OgIXZX5MU0/qOke2lnjLE7rwtZFdua+HstvPFGy49gHP6O3j12+24S/Zwel/HMR1Jz/9Qg8MGI+37qKioDGu8rRXu69sa0RQraLyhFvR4jTHN/gCXAS/6bV8DPO23bQOygQHe7WxgQkvPO378eNNWixYtMsYY8/rS7ab/PXPNvtLqNj3P11uLTP975prl+QfbHEsgfPGGU53Lba55aZnpf89c85MXl5r1BSXGGGPc7nrzxBdbzYB755o/fLrRGBMZ8bZGNMUbTbEao/GGWlviBVaaJvJqIC30AsB/8HEf7z6fFGAUkO1t9fUEPhaR2caYlW39oAlEvLfk0tb5XIqrPC38aK+hB8Jht/HitRN4bekOnlyYw3lPLmZwj2S6JsayfPshLh6bxV1nDQ13mEqpdggkoa8AhojIQDyJ/ArgKt9BY0wpkOHbFpFs4DehTubguVMUaPPQxZIqTw09mke5tEasw8YN0wZy6bgsPlhTwBebCtm8t5z/N3sk107pr+N5lYpyLSZ0Y4xLRG4DFgB24GVjzAYReQhP0//jUAfZFF+naFuHLhZXelro6Z2ghe4vPTGWn04dyE+n6opDSllJQLf+G2PmAfOO2vdAE+dOb39YgYn3ttDbulB0cZWTlDgHMfaovr9KKaWAqL9TtH0t9JIqJ+lReJeoUko1JsoTevtq6MVVdZYYg66UUhDtCb2h5NLGFnp1XcMSVkopFe2iOqE3DFtsxyiX9AQtuSilrCGqE3rDsMW2jkOvdHaKMehKqc4huhO6t4XellEuLnc9ZTUuLbkopSwjyhO6r1O09S300urOc5eoUqpz6LQJveG2/yRtoSulrCGqE7rDbsNhkzZ1ivpu+9eSi1LKKqI6oUPbF4ou8bbQdZSLUsoqoj6hexaKbn0LvbjKOotbKKUUWCChxzlsbRq22NBC11v/lVIWEf0JPcZOTZs6RZ04bEJKXEDzkymlVMSL/oTusFHbhnHoxVV1pCfG6BzgSinLiP6EHmNv07DFkiqnjnBRSllK9Cd0h63NnaI6wkUpZSWWSOhtHbaoLXSllJVYIKG3teRSp7f9K6UsJeoTenxM60suZTV17K+opVd6QoiiUkqpjhf1CT3OYW/1OPTvth3EXW84eVC3EEWllFIdL/oTehta6EtyD5AYa2dcvy4hikoppTpe1Cf0+Da00BfnHGDSwK7EOqL+n6+UUg2iPqN5WuiBJ/SCkmryDlQydXBGCKNSSqmOF/0J3WHD6a7HXW8COn9xzn4AThnSPZRhKaVUh7NAQvcsQ+cMsJX+Tc4BeqTEMTQzOZRhKaVUh4v6hB7vWyg6gI7R+nrDt9sOMm1whs7hopSynKhP6IcXim65hb67uJpDlU5OGtg11GEppVSHs0BCD7yFXl7rWxhab/lXSllP9Cf0mMAXiq6s9ST9ZJ0DXSllQVGf0OO9JZdAxqJXOl0AJMbZQxqTUkqFQ0AJXURmisgWEckVkXsbOX6riKwTkbUislhERgQ/1Mb5Wug1AZRcKms9CV1b6EopK2oxoYuIHXgGmAWMAK5sJGG/aYwZbYw5EfgL8LegR9qEuNa00L0JPTFWW+hKKesJpIU+Ecg1xuQZY5zAHOBC/xOMMWV+m0lAYHf5BEFrhi1qDV0pZWViTPO5V0QuA2YaY270bl8DTDLG3HbUeb8A7gJigdONMTmNPNfNwM0AmZmZ4+fMmdOmoCsqKkhO9twYtLu8nt8vqea/ToxjYs/mE/XH25y8n1PHi2cn4rB13Dh0/3ijgcYbOtEUK2i8odaWeGfMmLHKGDOh0YPGmGZ/gMuAF/22rwGebub8q4B/tfS848ePN221aNGiht/z91eY/vfMNf9etavFx/1p3iYz5L55bX7dtvKPNxpovKETTbEao/GGWlviBVaaJvJqICWXAqCv33Yf776mzAEuCuB5g6J1wxZdOsJFKWVZgST0FcAQERkoIrHAFcDH/ieIyBC/zfOAY8otoZIQ40nQ1c4AauhOF0mxWj9XSllTi9nNGOMSkduABYAdeNkYs0FEHsLT9P8YuE1EzgTqgGLgulAG7S/Rm6B9I1iaU1nrIklb6EopiwqouWqMmQfMO2rfA36/3xHkuAIW67ARa7dR4Ww5oVc53STpCBellEVF/Z2iAElxdqpqWy65VNRqyUUpZV0WSeiOgEouVbVuLbkopSzLGgk91kFFAAldW+hKKSuzRkKPszdMvNWcKqdLa+hKKcuySEJ3UBFADb2y1q3j0JVSlmWJhJ4cQA3d6arH6a4nWUsuSimLskRCD6RTtKphLnRN6Eopa7JEQg+khV7p9M20qCUXpZQ1WSKhezpF3b7JwRp1eC50baErpazJIgndgbveNDtBl65WpJSyOmskdG+ru7mx6L7FLXS1IqWUVVkjoce1PEGXb5y6jkNXSlmVJRK6r6Oz+Ra6JnSllLVZIqEfbqE3fXORb5SLzuWilLIqayX0Zm7/b2ih6ygXpZRFWSKhJwdQQ6+qdSGinaJKKeuyREIPpFO0otZNUqwDEemosJRSqkNZIqEnNwxbbKaGXuvS1rlSytIskdB9Myi2NGxRbypSSlmZJRJ6jN1GrMPWfEKvdenUuUopS7NEQgdPx2iz49Cdbh3hopSyNMsk9KQ4O1XO5mvoelORUsrKrJPQW1hXtMrp1oSulLI0yyT0luZE9ywQrTV0pZR1WSaht7RqUZWWXJRSFmeZhN5cp2h9vfF2imoLXSllXZZJ6Imx9iYn56qu803MpS10pZR1WSahN1dyaVh+ThO6UsrCLJPQk+McVDpdja4rqgtEK6U6g4ASuojMFJEtIpIrIvc2cvwuEdkoIj+IyEIR6R/8UJuXFOeg3kBN3bHriuoC0UqpzqDFhC4iduAZYBYwArhSREYcddoaYIIxZgzwHvCXYAfakuZWLdIFopVSnUEgLfSJQK4xJs8Y4wTmABf6n2CMWWSMqfJuLgX6BDfMljU3ha5v4QudbVEpZWWBNFmzgF1+27uBSc2cfwMwv7EDInIzcDNAZmYm2dnZgUV5lIqKimMem1/oSdpffbuU7alHJu4Vez3HNv6whtK8ju82aCzeSKbxhk40xQoab6gFPV5jTLM/wGXAi37b1wBPN3HuT/C00ONaet7x48ebtlq0aNEx+xbn7Df975lrlm47cMyxt5btMP3vmWt2F1e1+TXbo7F4I5nGGzrRFKsxGm+otSVeYKVpIq8G0kIvAPr6bffx7juCiJwJ3A+cZoypbcdnTJv4yimNrSvaMMpFO0WVUhYWSP1hBTBERAaKSCxwBfCx/wkiMhZ4HphtjCkKfpgtO7yu6LE3Fx0eh641dKWUdbWY0I0xLuA2YAGwCXjHGLNBRB4Skdne0x4FkoF3RWStiHzcxNOFTEudorEOGzF2ywy7V0qpYwRUgzDGzAPmHbXvAb/fzwxyXK3mS+hNDVvUeVyUUlZnmSarL2E3VnKpqtW50JVS1meZhO6w24iPsTXaKeqZC10TulLK2iyT0KHpKXQ9qxVpyUUpZW2WSuip8TGUVdcds79CF7dQSnUC1kroCTGUNpLQK7XkopTqBCyV0NMTG0/oVU63jkFXSlmepRJ6WhMt9Ipal860qJSyPEsl9PSEGEqqGmuhaw1dKWV9lkroaQkxlNXUUV9/eNWiWpebOrfRG4uUUpZnrYSeGIsxUF5zeOhiVa0uEK2U6hwsldDTE2IAKKl2NuzzjUvXUS5KKauzVEJP8yZ0/47RKqe20JVSnYOlEnp6oreF7tcxWqFT5yqlOglLJfTGW+i6QLRSqnOwVkL3tdD9EnrD4hY6ykUpZXHWSui+FnrV4U5R33S62kJXSlmdpRJ6nMNOQoz9iJKLbzrdRB3lopSyOEsldPB0jPp3imoLXSnVWVguoR89n0tlrQubQHyM5f6pSil1BMtlubSEmCM7RZ2eqXNFJIxRKaVU6FkyoZcd1ULXMehKqc7Acgn9mBq6UxeIVkp1DhZM6LFHzOWiqxUppToLyyX0tIQYaurqqanzjG6pqtUFopVSnYMlEzrQUEev0Ba6UqqTsGxC9w1d1NWKlFKdheUSevpR87lUaMlFKdVJWC6hH57Pxa+FriUXpVQnYLmEnp4QC3ha6PX1hiqnm0QtuSilOgHLJfSGKXSrnA0TcyVryUUp1QkElNBFZKaIbBGRXBG5t5Hjp4rIahFxichlwQ8zcClxDkQ8o1x8y8/pTItKqc6gxYQuInbgGWAWMAK4UkRGHHXaTuB64M1gB9haNps0zOfiW35OZ1pUSnUGgWS6iUCuMSYPQETmABcCG30nGGO2e4/VhyDGVvPNuFhV62uha8lFKWV9gST0LGCX3/ZuYFJbXkxEbgZuBsjMzCQ7O7stT0NFRUWzj7W5asjbXcjiZYcAyN28gdj9m9v0WsHQUryRRuMNnWiKFTTeUAt2vB1aizDGvAC8ADBhwgQzffr0Nj1PdnY2zT325bzl7CutZsjwYbBiJSdPHM+JfdPb9FrB0FK8kUbjDZ1oihU03lALdryBdIoWAH39tvt490Wss0ZksrWwgvnr9wE6ykUp1TkEktBXAENEZKCIxAJXAB+HNqz2uXxCX/p2TeDfq3cD6K3/SqlOocWEboxxAbcBC4BNwDvGmA0i8pCIzAYQkZNEZDfwI+B5EdkQyqBbEuuwcddZQxu2ddiiUqozCCjTGWPmAfOO2veA3+8r8JRiIsbsE7J4/qs8Nu8rJ0lHuSilOgHLNl3tNuHRy07gq61FOOyWuyFWKaWOYdmEDjC6Txqj+6SFOwyllOoQ2nRVSimL0ISulFIWoQldKaUsQhO6UkpZhCZ0pZSyCE3oSillEZrQlVLKIjShK6WURYgxJjwvLLIf2NHGh2cAB4IYTqhpvKEVTfFGU6yg8YZaW+Ltb4zp3tiBsCX09hCRlcaYCeGOI1Aab2hFU7zRFCtovKEW7Hi15KKUUhahCV0ppSwiWhP6C+EOoJU03tCKpnijKVbQeEMtqPFGZQ1dKaXUsaK1ha6UUuoomtCVUsoioi6hi8hMEdkiIrkicm+44zmaiPQVkUUislFENojIHd79XUXkcxHJ8f63S7hj9RERu4isEZG53u2BIrLMe43f9i4OHhFEJF1E3hORzSKySUSmRPi1vdP7d7BeRN4SkfhIur4i8rKIFInIer99jV5P8XjSG/cPIjIuQuJ91Pv38IOIfCAi6X7HfueNd4uInBMJ8fod+7WIGBHJ8G63+/pGVUIXETvwDDALGAFcKSIjwhvVMVzAr40xI4DJwC+8Md4LLDTGDAEWercjxR14FgD3+TPwd2PMYKAYuCEsUTXuCeA/xphhwAl44o7IaysiWcDtwARjzCjADlxBZF3fV4GZR+1r6nrOAoZ4f24G/tFBMfp7lWPj/RwYZYwZA2wFfgfgfd9dAYz0PuZZbw7pSK9ybLyISF/gbGCn3+72X19jTNT8AFOABX7bvwN+F+64Woj5I+AsYAvQy7uvF7Al3LF5Y+mD5017OjAXEDx3rjkau+ZhjjUNyMfbme+3P1KvbRawC+iKZ7nHucA5kXZ9gQHA+pauJ/A8cGVj54Uz3qOOXQy84f39iPwALACmREK8wHt4GiTbgYxgXd+oaqFz+A3is9u7LyKJyABgLLAMyDTG7PUe2gdkhimsoz0O/Bao9253A0qMMS7vdiRd44HAfuAVb4noRRFJIkKvrTGmAPgrnlbYXqAUWEXkXl+fpq5nNLz/fgbM9/4ekfGKyIVAgTHm+6MOtTveaEvoUUNEkoF/A78yxpT5HzOej9+wjxcVkfOBImPMqnDHEiAHMA74hzFmLFDJUeWVSLm2AN7a84V4Poh6A0k08vU7kkXS9WyJiNyPp+T5RrhjaYqIJAL3AQ+E4vmjLaEXAH39tvt490UUEYnBk8zfMMa8791dKCK9vMd7AUXhis/PVGC2iGwH5uApuzwBpIuIw3tOJF3j3cBuY8wy7/Z7eBJ8JF5bgDOBfGPMfmNMHfA+nmseqdfXp6nrGbHvPxG5HjgfuNr7IQSRGe8gPB/w33vfd32A1SLSkyDEG20JfQUwxDtKIBZPh8fHYY7pCCIiwEvAJmPM3/wOfQxc5/39Ojy19bAyxvzOGNPHGDMAz7X80hhzNbAIuMx7WkTECmCM2QfsEpHjvbvOADYSgdfWaycwWUQSvX8Xvngj8vr6aep6fgxc6x2NMRko9SvNhI2IzMRTNpxtjKnyO/QxcIWIxInIQDydjcvDEaOPMWadMaaHMWaA9323Gxjn/dtu//Xt6A6CIHQwnIunJ3sbcH+442kkvml4vqL+AKz1/pyLpza9EMgBvgC6hjvWo+KeDsz1/n4cnj/8XOBdIC7c8fnFeSKw0nt9PwS6RPK1Bf4fsBlYD7wGxEXS9QXewlPfr/Mmlxuaup54Osyf8b731uEZvRMJ8ebiqT373m/P+Z1/vzfeLcCsSIj3qOPbOdwp2u7rq7f+K6WURURbyUUppVQTNKErpZRFaEJXSimL0ISulFIWoQldKaUsQhO6UkpZhCZ0pZSyiP8Ph9T4Zl7lsuMAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light",
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.grid()\n",
        "plt.plot(np.arange(140), normal_train_data[0])\n",
        "plt.title(\"A Normal ECG\")\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "QpI9by2ZA0NN"
      },
      "source": [
        "Plot an anomalous ECG."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "zrpXREF2siBr",
        "outputId": "bb125d4c-4074-49d4-93ac-ad7120a4dd6c"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light",
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.grid()\n",
        "plt.plot(np.arange(140), anomalous_train_data[0])\n",
        "plt.title(\"An Anomalous ECG\")\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "0DS6QKZJslZz"
      },
      "source": [
        "### Picking an Embedding to Build the Model\n",
        "\n",
        "After training and evaluating the example model, try modifying the size and number of layers to build an understanding for autoencoder architectures.\n",
        "\n",
        "Note: Changing the size of the embedding (the smallest layer) can produce interesting results. Feel free to play around with that layer size."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "bf6owZQDsp9y"
      },
      "outputs": [],
      "source": [
        "EMBEDDING_SIZE = # YOUR CODE GOES HERE #\n",
        "\n",
        "class AnomalyDetector(Model):\n",
        "  def __init__(self):\n",
        "    super(AnomalyDetector, self).__init__()\n",
        "    self.encoder = tf.keras.Sequential([\n",
        "      layers.Dense(8, activation=\"relu\"),\n",
        "      layers.Dense(EMBEDDING_SIZE, activation=\"relu\")]) # Smallest Layer Defined Here\n",
        "    \n",
        "    self.decoder = tf.keras.Sequential([\n",
        "      layers.Dense(8, activation=\"relu\"),\n",
        "      layers.Dense(140, activation=\"sigmoid\")])\n",
        "    \n",
        "  def call(self, x):\n",
        "    encoded = self.encoder(x)\n",
        "    decoded = self.decoder(encoded)\n",
        "    return decoded\n",
        "\n",
        "autoencoder = AnomalyDetector()\n",
        "print(\"Chosen Embedding Size: \", EMBEDDING_SIZE)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "gwRpBBbg463S"
      },
      "outputs": [],
      "source": [
        "optimizer = tf.keras.optimizers.Adam(learning_rate=0.01)\n",
        "autoencoder.compile(optimizer=optimizer, loss='mae')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "zuTy60STBEy4"
      },
      "source": [
        "### Train the model\n",
        "\n",
        "Notice that the autoencoder is now trained using the combined training data which is primarily normal ECGs with some anomalies mixed in. It is still evaluated using the full test set."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "V6NFSs-jsty2"
      },
      "outputs": [],
      "source": [
        "history = autoencoder.fit(combined_train_data, combined_train_data, epochs=50, \n",
        "                          batch_size=512, validation_data=(test_data, test_data), shuffle=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OEexphFwwTQS"
      },
      "outputs": [],
      "source": [
        "plt.plot(history.history[\"loss\"], label=\"Training Loss\")\n",
        "plt.plot(history.history[\"val_loss\"], label=\"Validation Loss\")\n",
        "plt.legend()\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "ceI5lKv1BT-A"
      },
      "source": [
        "### Evaluate Training\n",
        "\n",
        "You will soon classify an ECG as anomalous if the reconstruction error is greater than one standard deviation from the normal training examples. First, let's plot a normal ECG from the training set, the reconstruction after it's encoded and decoded by the autoencoder, and the reconstruction error."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "hmsk4DuktxJ2",
        "outputId": "f4a843db-9ea5-4e04-879a-4a5a91dfd4c3"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light",
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "encoded_imgs = autoencoder.encoder(normal_test_data).numpy()\n",
        "decoded_imgs = autoencoder.decoder(encoded_imgs).numpy()\n",
        "\n",
        "plt.plot(normal_test_data[0],'b')\n",
        "plt.plot(decoded_imgs[0],'r')\n",
        "plt.fill_between(np.arange(140), decoded_imgs[0], normal_test_data[0], color='lightcoral' )\n",
        "plt.legend(labels=[\"Input\", \"Reconstruction\", \"Error\"])\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "ocA_q9ufB_aF"
      },
      "source": [
        "Create a similar plot, this time for an anomalous test example."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "vNFTuPhLwTBn",
        "outputId": "4f4dc927-3edb-4b8a-c0de-9b03b23e011e"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light",
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "encoded_imgs = autoencoder.encoder(anomalous_test_data).numpy()\n",
        "decoded_imgs = autoencoder.decoder(encoded_imgs).numpy()\n",
        "\n",
        "plt.plot(anomalous_test_data[0],'b')\n",
        "plt.plot(decoded_imgs[0],'r')\n",
        "plt.fill_between(np.arange(140), decoded_imgs[0], anomalous_test_data[0], color='lightcoral' )\n",
        "plt.legend(labels=[\"Input\", \"Reconstruction\", \"Error\"])\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "N2sl_2z-uOv3"
      },
      "source": [
        "### ROC and AUC Metrics\n",
        "\n",
        "The Receiver Operating Characteristic (ROC) plots allows us to visualize the tradeoff between predicting anomalies as normal (false positives) and predicting normal data as an anomaly (false negative). Normal rhythms are labeled as `1` in this dataset but we have to flip them here to match the ROC curves expectations.\n",
        "\n",
        "The ROC plot now has threshold values plotted on their corrispoinding points on the curve to aid in selecting a theshold for the application."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "LketUNanyLte",
        "outputId": "77025e7d-9066-4095-e76d-c7f8c39b6d39"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light",
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "reconstructions = autoencoder(test_data)\n",
        "loss = tf.keras.losses.mae(reconstructions, test_data)\n",
        "fpr = []\n",
        "tpr = []\n",
        "#the test labels are flipped to match how the roc_curve function expects them.\n",
        "flipped_labels = 1-test_labels \n",
        "fpr, tpr, thresholds = roc_curve(flipped_labels, loss)\n",
        "plt.figure()\n",
        "lw = 2\n",
        "plt.plot(fpr, tpr, color='darkorange',\n",
        "         lw=lw, label='ROC curve ')\n",
        "plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n",
        "plt.xlim([0.0, 1.0])\n",
        "plt.ylim([0.0, 1.05])\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.title('Receiver operating characteristic example')\n",
        "plt.legend(loc=\"lower right\")\n",
        "\n",
        "# plot some thresholds\n",
        "thresholds_every=20\n",
        "thresholdsLength = len(thresholds)\n",
        "colorMap=plt.get_cmap('jet', thresholdsLength)\n",
        "for i in range(0, thresholdsLength, thresholds_every):\n",
        "  threshold_value_with_max_four_decimals = str(thresholds[i])[:5]\n",
        "  plt.scatter(fpr[i], tpr[i], c='black')\n",
        "  plt.text(fpr[i] - 0.03, tpr[i] + 0.005, threshold_value_with_max_four_decimals, fontdict={'size': 15});\n",
        "\n",
        "plt.show()\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "Hk6LfUhB1ohP"
      },
      "source": [
        "Since our model does a great job in diferentiating normal rythms from abnormal ones it seems easy to pick the threshold that would give us the high true positive rate (TPR) and low false positive rate (FPR) that is at the 'knee' of the curve.\n",
        "\n",
        "However, in some cases there may be an application constraint that requires a specific TPR or FPR, in which case we would have to move off of the 'knee' and sacrifice overall accuracy. In this case we might rather have false alarms than miss a potentially dangerous rythm."
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "6sH6Yv-kx17h"
      },
      "source": [
        "Now that we understand how to visualize the impact of the selected threshold, we calculate the area under the ROC curve (AUC). \n",
        "\n",
        "This metric is very useful for evalutation of a specfic model design. Adjust the size of the encoding layer (smallest layer) in the autoencoder to maximize this metric."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "qDMEZfCvyoaP"
      },
      "outputs": [],
      "source": [
        "roc_auc = auc(fpr, tpr)\n",
        "print(roc_auc)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "__CBKkwzhGkw"
      },
      "source": [
        "### Picking a Threshold to Detect Anomalies\n",
        "Detect anomalies by calculating whether the reconstruction loss is greater than a fixed threshold.\n",
        "\n",
        "Try to maximize the accuracy, precision, and recall. Think about the application and the consequences of a false positive and a false negative. The course staff was able to acheive greater than 94% on all three at the same time!\n",
        "\n",
        "[More details on precision and recall](https://developers.google.com/machine-learning/crash-course/classification/precision-and-recall)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fi7mAGvzhGkx"
      },
      "outputs": [],
      "source": [
        "threshold = # YOUR CODE GOES HERE #\n",
        "print(\"Chosen Threshold: \", threshold)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "id": "lKRN0GBchGkx"
      },
      "outputs": [],
      "source": [
        "def predict(model, data, threshold):\n",
        "  reconstructions = model(data)\n",
        "  loss = tf.keras.losses.mae(reconstructions, data)\n",
        "  return tf.math.less(loss, threshold), loss\n",
        "\n",
        "def print_stats(predictions, labels):\n",
        "  print(\"Accuracy = {}\".format(accuracy_score(labels, predictions)))\n",
        "  print(\"Precision = {}\".format(precision_score(labels, predictions)))\n",
        "  print(\"Recall = {}\".format(recall_score(labels, predictions)))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xW6o5LoXhGkx"
      },
      "outputs": [],
      "source": [
        "preds, scores = predict(autoencoder, test_data, threshold)\n",
        "print_stats(preds, test_labels)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "name": "3-8-16-Assignment.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}