4.Linear Regression (multiple variable).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "assignment04.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sQy9EvTe1J6q",
        "colab_type": "text"
      },
      "source": [
        "### **0. Preparation**\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YgZQmLLx1HWI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import copy\n",
        "import math\n",
        "import csv\n",
        "from collections import deque\n",
        "from google.colab import files\n",
        "from mpl_toolkits.mplot3d import Axes3D\n",
        "from matplotlib import cm\n",
        "from matplotlib.ticker import LinearLocator, FormatStrFormatter"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oDTD9ppr1OAX",
        "colab_type": "text"
      },
      "source": [
        "### **1. Input Data**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XjbPUG9E1Tie",
        "colab_type": "code",
        "outputId": "acb84d08-ea2e-4839-c07d-4d5123be6e42",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 113
        }
      },
      "source": [
        "uploaded = files.upload()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-cc2da0a6-36d8-4fc4-a644-57d2eb67ec13\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-cc2da0a6-36d8-4fc4-a644-57d2eb67ec13\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving data_test.csv to data_test (1).csv\n",
            "Saving data_train.csv to data_train (1).csv\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MFaPtak-zpwW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_data = deque()\n",
        "test_data = deque()\n",
        "\n",
        "with open('data_train.csv', newline='') as myfile:\n",
        "    reader  = csv.reader(myfile, delimiter=',')\n",
        "    ct = 1 \n",
        "    for i in reader:\n",
        "      train_data.append([float(i[0]), float(i[1]), float(i[2]), float(i[3])])\n",
        "      #print('[', ct, ']', 'x =', i[0], ', y = ', i[1], ', z = ', i[2], ', h = ', i[3])\n",
        "      ct += 1\n",
        "\n",
        "with open('data_test.csv', newline='') as myfile:\n",
        "    reader  = csv.reader(myfile, delimiter=',')\n",
        "    ct = 1 \n",
        "    for i in reader:\n",
        "      test_data.append([float(i[0]), float(i[1]), float(i[2]), float(i[3])])\n",
        "      #print('[', ct, ']', 'x =', i[0], ', y = ', i[1], ', z = ', i[2], ', h = ', i[3])\n",
        "      ct += 1\n",
        "\n",
        "#print(type(train_data[299][1]))\n",
        "#print(test_data[299])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Cw4DKfx74hAg",
        "colab_type": "text"
      },
      "source": [
        "### **2. Linear Regression**\n",
        "1) Linear Model\n",
        "<br>\n",
        "$\\ f_\\theta(x,y,z) = \\theta_0 + \\theta_1x + \\theta_2y + \\theta_3z  $\n",
        "<br>\n",
        "<br>\n",
        "2) Objective Function\n",
        "<br>\n",
        "$\\ J(\\theta_0,\\theta_1,\\theta_2,\\theta_3 ) = \\frac{1}{2m} \\sum_{i=1}^{m}(\\theta_0 + \\theta_1x^{(i)} + \\theta_2y^{(i)} + \\theta_3z^{(i)} - h^{(i)})^2 $\n",
        "<br>\n",
        "<br>\n",
        "3) Gradient Descent\n",
        "<br>\n",
        "$\\ \\theta_0^{(t+1)} =  \\theta_0^{(t)} - \\alpha \\frac{1}{m} \\sum_{i=1}^{m} (f_\\theta(x^{(i)},y^{(i)},z^{(i)}) - h^{(i)})$\n",
        "<br>\n",
        "$\\ \\theta_1^{(t+1)} = \\theta_0^{(t)} - \\alpha \\frac{1}{m} \\sum_{i=1}^{m} (f_\\theta(x^{(i)},y^{(i)},z^{(i)}) - h^{(i)})x^{(i)}$\n",
        "<br>\n",
        "$\\ \\theta_2^{(t+1)} = \\theta_0^{(t)} - \\alpha \\frac{1}{m} \\sum_{i=1}^{m} (f_\\theta(x^{(i)},y^{(i)},z^{(i)}) - h^{(i)})y^{(i)}$\n",
        "<br>\n",
        "$\\ \\theta_3^{(t+1)} = \\theta_0^{(t)} - \\alpha \\frac{1}{m} \\sum_{i=1}^{m} (f_\\theta(x^{(i)},y^{(i)},z^{(i)}) - h^{(i)})z^{(i)}$\n",
        "<br>\n",
        "<br>\n",
        "4) HyperParameters Setting\n",
        "<br>\n",
        "$\\ \\theta_0^{(0)} = -10$  \n",
        "$\\ \\theta_1^{(0)} = -20$  \n",
        "$\\ \\theta_2^{(0)} = 10$  \n",
        "$\\ \\theta_3^{(0)} = 20$\n",
        "<br>\n",
        "$\\ lr = 3*10^{-4}$\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3-gCevCjgKyl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def objectiveFunction(data, theta_0, theta_1, theta_2, theta_3):\n",
        "  cost = 0\n",
        "  predictions = deque()\n",
        "  for i in range(0, 300):\n",
        "    predictions.append(theta_0 + theta_1 * data[i][0] + theta_2 * data[i][1] + theta_3 * data[i][2] - data[i][3])\n",
        "  cost = np.sum(np.square(predictions)) / 600\n",
        "  #print(\"Cost\",i,\"번째\",cost)\n",
        "  return cost"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NsaVBnlHiAys",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def gradientDescent(data, theta_0, theta_1, theta_2, theta_3):\n",
        "  for i in range(0, 300):\n",
        "    lr = 0.0003\n",
        "    estimate = theta_0 + theta_1 * data[i][0] + theta_2 * data[i][1] + theta_3 * data[i][2] \n",
        "    theta_0 -= lr * (estimate - data[i][3]) / 300\n",
        "    theta_1 -= lr * data[i][0] * (estimate - data[i][3]) / 300\n",
        "    theta_2 -= lr * data[i][1] * (estimate - data[i][3]) / 300\n",
        "    theta_3 -= lr * data[i][2] * (estimate - data[i][3]) / 300\n",
        "  return theta_0, theta_1, theta_2, theta_3"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nKROWnxQlgsK",
        "colab_type": "code",
        "outputId": "961496cc-b4fc-4e2b-8b3b-67029e008002",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 857
        }
      },
      "source": [
        "def main():\n",
        "  theta_0,theta_1,theta_2,theta_3 = -10, -20, 10, 20  #Initialization\n",
        "  train_costs = deque()\n",
        "  test_costs = deque()\n",
        "  theta_0s = deque()\n",
        "  theta_1s = deque()\n",
        "  theta_2s = deque()\n",
        "  theta_3s = deque()\n",
        "  iters = deque()\n",
        "\n",
        "  for iter in range(0, 100000):\n",
        "    train_value = objectiveFunction(train_data, theta_0, theta_1, theta_2, theta_3)\n",
        "    theta_0, theta_1, theta_2, theta_3 = gradientDescent(train_data, theta_0, theta_1, theta_2, theta_3)\n",
        "    test_value = objectiveFunction(test_data, theta_0, theta_1, theta_2, theta_3)\n",
        "\n",
        "    #data 수집\n",
        "    iters.append(iter)\n",
        "    train_costs.append(train_value)\n",
        "    test_costs.append(test_value)\n",
        "    theta_0s.append(theta_0)\n",
        "    theta_1s.append(theta_1)\n",
        "    theta_2s.append(theta_2)\n",
        "    theta_3s.append(theta_3)\n",
        "\n",
        "  print(\"Plotting the Model Parameters\")\n",
        "  fig = plt.figure()\n",
        "  graph1 = plt.plot(iters, theta_0s, 'k')\n",
        "  graph2 = plt.plot(iters, theta_1s, 'r')\n",
        "  graph3 = plt.plot(iters, theta_2s, 'g')\n",
        "  graph4 = plt.plot(iters, theta_3s, 'b')\n",
        "  plt.xlabel('iterations')\n",
        "  plt.ylabel('value')\n",
        "  plt.show()\n",
        "\n",
        "  print(\"Plotting the Training Errors\")\n",
        "  plt.plot(iters, train_costs, 'b')\n",
        "  plt.xlabel('iteration')\n",
        "  plt.ylabel('Traing Errors')\n",
        "  plt.show()\n",
        "\n",
        "  print(\"Plotting the Test Errors\")\n",
        "  plt.plot(iters, test_costs, 'r')\n",
        "  plt.xlabel('iteration')\n",
        "  plt.ylabel('Test Errors')\n",
        "  plt.show()\n",
        "\n",
        "if __name__ == \"__main__\":\n",
        "  main()"
      ],
      "execution_count": 84,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Plotting the Model Parameters\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Plotting the Training Errors\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZsAAAEGCAYAAACzYDhlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAcjklEQVR4nO3df5QddZnn8ffHxCALQoK0OdkkmLBm9UTOiKGFsLoehDEExjXsHldhXJNB1uiAv5aZdcJ49mSE9SyOc/yR1YlGiSSKICJKlgEz2QjjjJ5AOoD8lEkDYjoLJCYQBthFwGf/qKel6LldfZPub9/k5vM6555b9dS36vstqtMPVfV0lSICMzOzkl7W6QGYmVn3c7IxM7PinGzMzKw4JxszMyvOycbMzIqb2OkB7C+OPvromDVrVqeHYWZ2QNmyZcuvI6JnpHZONmnWrFn09fV1ehhmZgcUSQ+3086X0czMrDgnGzMzK87JxszMinOyMTOz4pxszMysOCcbMzMrzsnGzMyKc7IZpW9/G7761U6Pwsxs/+ZkM0pXXgmXXdbpUZiZ7d+cbMzMrDgnGzMzK87JxszMinOyMTOz4pxszMysOCebMRDR6RGYme3fiiUbSa+TdEft86SkT0g6StIGSVvze0q2l6QVkvol3SlpXm1bS7L9VklLavETJN2V66yQpIy37KPMfpbasplZ9yiWbCLi/og4PiKOB04AngF+ACwDNkbEHGBjzgOcAczJz1JgJVSJA1gOnAScCCyvJY+VwAdr6y3M+HB9mJlZB4zXZbTTgAci4mFgEbAm42uAs3J6EbA2KpuAyZKmAacDGyJid0Q8DmwAFuayIyJiU0QEsHbItlr1YWZmHTBeyeZs4MqcnhoRj+T0o8DUnJ4ObKutM5CxpvhAi3hTH2Zm1gHFk42kScC7gO8NXZZnJEVvrzf1IWmppD5JfTt37iw5DDOzg9p4nNmcAdwWEY/l/GN5CYz83pHx7cDM2nozMtYUn9Ei3tTHS0TEqojojYjenp6efdw9MzMbyXgkm3N48RIawDpgsKJsCXBdLb44q9LmA3vyUth6YIGkKVkYsABYn8uelDQ/q9AWD9lWqz6KcOmzmVmziSU3Lukw4B3Ah2rhS4GrJZ0HPAy8J+M3AGcC/VSVa+cCRMRuSZcAm7PdxRGxO6fPBy4HDgVuzE9TH2POpc9mZiMrmmwi4mngVUNiu6iq04a2DeCCYbazGljdIt4HHNci3rIPMzPrDD9BwMzMinOyMTOz4pxszMysOCebMeBqNDOzZk42o+RqNDOzkTnZmJlZcU42ZmZWnJONmZkV52RjZmbFOdmYmVlxTjZjwKXPZmbNnGxGyaXPZmYjc7IxM7PinGzMzKw4JxszMyvOycbMzIpzsjEzs+KcbMaAS5/NzJoVTTaSJku6RtIvJN0n6WRJR0naIGlrfk/JtpK0QlK/pDslzattZ0m23yppSS1+gqS7cp0VUlWIPFwfZfax1JbNzLpH6TObLwE/iojXA28E7gOWARsjYg6wMecBzgDm5GcpsBKqxAEsB04CTgSW15LHSuCDtfUWZny4PszMrAOKJRtJRwJvAy4DiIjfRMQTwCJgTTZbA5yV04uAtVHZBEyWNA04HdgQEbsj4nFgA7Awlx0REZsiIoC1Q7bVqg8zM+uAkmc2s4GdwDcl3S7pG5IOA6ZGxCPZ5lFgak5PB7bV1h/IWFN8oEWchj5eQtJSSX2S+nbu3Lkv+2hmZm0omWwmAvOAlRHxJuBphlzOyjOSorfXm/qIiFUR0RsRvT09PSWHYWZ2UCuZbAaAgYi4JeevoUo+j+UlMPJ7Ry7fDsysrT8jY03xGS3iNPRhZmYdUCzZRMSjwDZJr8vQacC9wDpgsKJsCXBdTq8DFmdV2nxgT14KWw8skDQlCwMWAOtz2ZOS5mcV2uIh22rVRxEufTYzazax8PY/ClwhaRLwIHAuVYK7WtJ5wMPAe7LtDcCZQD/wTLYlInZLugTYnO0ujojdOX0+cDlwKHBjfgAuHaaPMefSZzOzkRVNNhFxB9DbYtFpLdoGcMEw21kNrG4R7wOOaxHf1aoPMzPrDD9BwMzMinOyMTOz4pxszMysOCcbMzMrzslmDLj02cysmZPNKLn02cxsZE42ZmZWnJONmZkV52RjZmbFOdmYmVlxTjZjwNVoZmbNnGxGydVoZmYjc7IxM7PinGzMzKw4JxszMyvOycbMzIpzsjEzs+KcbMaAS5/NzJoVTTaSfinpLkl3SOrL2FGSNkjamt9TMi5JKyT1S7pT0rzadpZk+62SltTiJ+T2+3NdNfVRZh9LbdnMrHuMx5nN2yPi+IjozfllwMaImANszHmAM4A5+VkKrIQqcQDLgZOAE4HlteSxEvhgbb2FI/RhZmYd0InLaIuANTm9BjirFl8blU3AZEnTgNOBDRGxOyIeBzYAC3PZERGxKSICWDtkW636MDOzDiidbAL4W0lbJC3N2NSIeCSnHwWm5vR0YFtt3YGMNcUHWsSb+ngJSUsl9Unq27lz517vnJmZtWdi4e2/NSK2S3o1sEHSL+oLIyIkFb293tRHRKwCVgH09vb6Nr+ZWSFFz2wiYnt+7wB+QHXP5bG8BEZ+78jm24GZtdVnZKwpPqNFnIY+zMysA4olG0mHSXrl4DSwALgbWAcMVpQtAa7L6XXA4qxKmw/syUth64EFkqZkYcACYH0ue1LS/KxCWzxkW636KMKlz2ZmzUpeRpsK/CCrkScC34mIH0naDFwt6TzgYeA92f4G4EygH3gGOBcgInZLugTYnO0ujojdOX0+cDlwKHBjfgAuHaaPMefSZzOzkY2YbCT9K2AgIp6VdArwe1RVY080rRcRDwJvbBHfBZzWIh7ABcNsazWwukW8Dziu3T7MzKwz2rmM9n3gBUmvpbqZPhP4TtFRmZlZV2kn2fw2Ip4H/j3wPyPivwLTyg7LzMy6STvJ5jlJ51DdaL8+Yy8vNyQzM+s27SSbc4GTgc9ExEOSZgPfKjssMzPrJo0FApImAJ+KiPcNxiLiIeCzpQd2IHHps5lZs8Yzm4h4AXiNpEnjNJ4DjkufzcxG1s7f2TwI/FTSOuDpwWBEfL7YqMzMrKu0k2weyM/LgFeWHY6ZmXWjEZNNRHwaQNLhOf9U6UGZmVl3GbEaTdJxkm4H7gHuydcFvKH80MzMrFu0U/q8CrgwIl4TEa8B/gT4etlhHVhcjWZm1qydZHNYRNw0OBMRNwOHFRuRmZl1nbaq0ST9N178Q87/RFWhZrj02cysHe2c2XwA6AGupXoo59EZMzMza0s7TxC4NiLePk7jMTOzLtTOEwR+K+nIcRqPmZl1oXbu2TwF3CVpAy99gsDHio3KzMy6SjvJ5tr82DBc+mxm1qzxMlres/mjiFgz9NNuB5ImSLpd0vU5P1vSLZL6JX138CGfkg7J+f5cPqu2jYsyfr+k02vxhRnrl7SsFm/ZRwmuRjMzG9l43LP5OHBfbf6zwBci4rXA48B5GT8PeDzjX8h2SJoLnA28AVgI/HUmsAnAV4AzgLnAOdm2qQ8zM+uAdkqfB+/ZXCZpxeCnnY1LmgH8AfCNnBdwKnBNNlkDnJXTi3KeXH5atl8EXBURz+a7dPqBE/PTHxEPRsRvgKuARSP0YWZmHVD6ns0XgU/y4tOiXwU8ERHP5/wAMD2npwPbACLieUl7sv10YFNtm/V1tg2JnzRCHy8haSmwFOCYY47Zh90zM7N2DJtsJB0REU+2uj8jacTfzJLeCeyIiC2SThndMMuIiFVUz36jt7fXt/nNzAppuox28+CEpI1Dlv2wjW2/BXiXpF9SXeI6FfgSMFnSYJKbAWzP6e3AzOxvInAksKseH7LOcPFdDX2YmVkHNCWbep3VUQ3LWoqIiyJiRkTMorrB/+OIeB9wE/DubLYEuC6n1+U8ufzHEREZPzur1WYDc4Bbgc3AnKw8m5R9rMt1huujCJc+m5k1a0o2Mcx0q/m98WfAhZL6qe6vXJbxy4BXZfxCYBlARNwDXA3cC/wIuCAiXsh7Mh8B1lNVu12dbZv6GHMufTYzG1lTgcCrJV1IdRYzOE3O9+xNJ/lagptz+kGqSrKhbf4f8B+HWf8zwGdaxG8AbmgRb9mHmZl1RlOy+TovVpHVpyFLmc3MzNoxbLKJiE+P50DMzKx7tfNHnWZmZqPiZGNmZsU52YwBlz6bmTUb8XE1tSq0uj3Aloi4Y+yHdGBx6bOZ2cjaObPpBT5M9Xyx6cCHqJ6+/HVJnyw4NjMz6xLtPIhzBjAvIp4CkLQc+BvgbcAW4C/LDc/MzLpBO2c2rwaerc0/B0yNiP87JG5mZtZSO2c2VwC3SBp8vti/A74j6TCqR8iYmZk1GjHZRMQlkn4E/JsMfTgi+nL6fcVGZmZmXaOdMxuA26ge0z8RqvfZRMSvio3qAOPSZzOzZu2UPn8UWA48BrxA9SDOAH6v7NAODC59NjMbWTtnNh8HXhcRu0oPxszMulM71WjbqP6I08zMbJ+0c2bzIHCzpL+hVuocEZ8vNiozM+sq7SSbX+VnUn7MzMz2Sjulz/v0XhtJrwB+AhyS/VwTEcslzQauonpd8xbg/RHxG0mHAGuBE4BdwHsj4pe5rYuA86gKFD4WEeszvhD4EjAB+EZEXJrxln3sy360w9VoZmbNhr1nI+mL+f2/JK0b+mlj288Cp0bEG4HjgYWS5gOfBb4QEa8FHqdKIuT34xn/QrZD0lzgbOANVM9k+2tJEyRNAL4CnAHMBc7JtjT0MeZcjWZmNrKmM5tv5fdf7cuGIyKAp3L25fkJ4FTgDzO+BvgLYCWwKKcBrgG+LEkZvyoingUektQPnJjt+iPiQQBJVwGLJN3X0IeZmXVA02uht+T33+3rxvPsYwvwWqqzkAeAJyLi+WwyQPUkafJ7W/b5vKQ9VJfBpgObaputr7NtSPykXGe4PszMrAPa+aPOOcD/oLpU9YrBeEQcO9K6EfECcLykycAPgNfv+1DHnqSlwFKAY445psOjMTPrXu38nc03qS5BPQ+8neom/rf3ppOIeAK4CTgZmCxpMMnNoHoMDvk9EyCXH0lVKPC7+JB1hovvauhj6LhWRURvRPT29PTszS6ZmdleaCfZHBoRGwFFxMMR8RfAH4y0kqSePKNB0qHAO4D7qJLOu7PZEmDwadLrcp5c/uO877MOOFvSIVllNge4FdgMzJE0W9IkqiKCdbnOcH2YmVkHtPN3Ns9KehmwVdJHqM4SDm9jvWnAmrxv8zLg6oi4XtK9wFWS/jtwO3BZtr8M+FYWAOymSh5ExD2SrqZ6ncHzwAV5eY4cz3qq0ufVEXFPbuvPhumjCJc+m5k1U4zwm1LSm6nOSCYDlwBHAJ+LiE2NKx5gent7o6+vb+SGQ7z//fCzn8EDDxQYlJnZfk7SlojoHald45lNnpW8NyL+lKqM+dwxGp+ZmR1Emv6oc2JernrrOI7HzMy6UNOZza3APOD2fGLA94CnBxdGxLWFx2ZmZl2inQKBV1CVE59K9QSAwZenOdmYmVlbmpLNqyVdCNzNi0lmkOuvzMysbU3JZgJViXOrR0062dS49NnMrFlTsnkkIi4et5EcoPzUZzOzkTU9QcC/Rs3MbEw0JZvTxm0UZmbW1YZNNhGxezwHYmZm3audB3GamZmNipONmZkV52QzBlz6bGbWzMlmlFz6bGY2MicbMzMrzsnGzMyKc7IxM7PinGzMzKy4YslG0kxJN0m6V9I9kj6e8aMkbZC0Nb+nZFySVkjql3SnpHm1bS3J9lslLanFT5B0V66zQqpu1w/Xh5mZdUbJM5vngT+JiLnAfOACSXOBZcDGiJgDbMx5gDOAOflZCqyEKnEAy4GTgBOB5bXksRL4YG29hRkfro8iXPpsZtasWLKJiEci4rac/ifgPmA6sAhYk83WAGfl9CJgbVQ2AZMlTQNOBzZExO6IeBzYACzMZUdExKaICGDtkG216mPMufTZzGxk43LPRtIs4E3ALcDUiHgkFz0KTM3p6cC22moDGWuKD7SI09DH0HEtldQnqW/nzp17v2NmZtaW4slG0uHA94FPRMST9WV5RlL0IlRTHxGxKiJ6I6K3p6en5DDMzA5qRZONpJdTJZorIuLaDD+Wl8DI7x0Z3w7MrK0+I2NN8Rkt4k19mJlZB5SsRhNwGXBfRHy+tmgdMFhRtgS4rhZfnFVp84E9eSlsPbBA0pQsDFgArM9lT0qan30tHrKtVn2YmVkHNL0WerTeArwfuEvSHRn7c+BS4GpJ5wEPA+/JZTcAZwL9wDPAuVC9V0fSJcDmbHdx7V075wOXA4cCN+aHhj6KcDWamVmzYskmIv6B4V8t/c/eApr3Vi4YZlurgdUt4n3AcS3iu1r1UYKr0czMRuYnCJiZWXFONmZmVpyTjZmZFedkY2ZmxTnZmJlZcU42Y8Clz2ZmzZxsRsmlz2ZmI3OyMTOz4pxszMysOCcbMzMrzsnGzMyKc7IxM7PinGzGgEufzcyaOdmMkkufzcxG5mRjZmbFOdmYmVlxTjZmZlack42ZmRVXLNlIWi1ph6S7a7GjJG2QtDW/p2RcklZI6pd0p6R5tXWWZPutkpbU4idIuivXWSFVt+qH68PMzDqn5JnN5cDCIbFlwMaImANszHmAM4A5+VkKrIQqcQDLgZOAE4HlteSxEvhgbb2FI/RRjEufzcyaFUs2EfETYPeQ8CJgTU6vAc6qxddGZRMwWdI04HRgQ0TsjojHgQ3Awlx2RERsiogA1g7ZVqs+inDps5nZyMb7ns3UiHgkpx8Fpub0dGBbrd1AxpriAy3iTX38M5KWSuqT1Ldz58592B0zM2tHxwoE8oyk6AWokfqIiFUR0RsRvT09PSWHYmZ2UBvvZPNYXgIjv3dkfDsws9ZuRsaa4jNaxJv6MDOzDhnvZLMOGKwoWwJcV4svzqq0+cCevBS2HlggaUoWBiwA1ueyJyXNzyq0xUO21aoPMzPrkImlNizpSuAU4GhJA1RVZZcCV0s6D3gYeE82vwE4E+gHngHOBYiI3ZIuATZnu4sjYrDo4HyqirdDgRvzQ0MfZmbWIcWSTUScM8yi01q0DeCCYbazGljdIt4HHNcivqtVHyW59NnMrJmfIDBKLn02MxuZk42ZmRXnZGNmZsU52ZiZWXFONmZmVpyTzRhwNZqZWTMnm1FyNZqZ2cicbMzMrDgnGzMzK87JxszMinOyMTOz4pxszMysOCebMeDSZzOzZk42o+TSZzOzkTnZmJlZcU42ZmZWnJONmZkV52RjZmbFdW2ykbRQ0v2S+iUt6/R4zMwOZl2ZbCRNAL4CnAHMBc6RNLdEX5MmwdNPu/zZzKzJxE4PoJATgf6IeBBA0lXAIuDese7opJPgy1+GmTPh8MP331Lo/XVcZtZ5118Pxx5bto9uTTbTgW21+QHgpKGNJC0FlgIcc8wx+9TROefAjh1w223w3HP7tInifNZlZk0OOaR8H92abNoSEauAVQC9vb379Ct5wgS48MIxHZaZWdfpyns2wHZgZm1+RsbMzKwDujXZbAbmSJotaRJwNrCuw2MyMztodeVltIh4XtJHgPXABGB1RNzT4WGZmR20ujLZAETEDcANnR6HmZl172U0MzPbjzjZmJlZcU42ZmZWnJONmZkVp/CflwMgaSfw8D6ufjTw6zEczoHA+3xw8D53v9Hu72siomekRk42Y0BSX0T0dnoc48n7fHDwPne/8dpfX0YzM7PinGzMzKw4J5uxsarTA+gA7/PBwfvc/cZlf33PxszMivOZjZmZFedkY2ZmxTnZjJKkhZLul9QvaVmnx7M3JM2UdJOkeyXdI+njGT9K0gZJW/N7SsYlaUXu652S5tW2tSTbb5W0pBY/QdJduc4Kaf94QbWkCZJul3R9zs+WdEuO87v5agokHZLz/bl8Vm0bF2X8fkmn1+L73c+EpMmSrpH0C0n3STq524+zpP+SP9d3S7pS0iu67ThLWi1ph6S7a7Hix3W4PhpFhD/7+KF6fcEDwLHAJODnwNxOj2svxj8NmJfTrwT+EZgL/CWwLOPLgM/m9JnAjYCA+cAtGT8KeDC/p+T0lFx2a7ZVrntGp/c7x3Uh8B3g+py/Gjg7p78K/HFOnw98NafPBr6b03PzeB8CzM6fgwn7688EsAb4zzk9CZjczceZ6tXwDwGH1o7vH3XbcQbeBswD7q7Fih/X4fpoHGun/xEcyB/gZGB9bf4i4KJOj2sU+3Md8A7gfmBaxqYB9+f014Bzau3vz+XnAF+rxb+WsWnAL2rxl7Tr4H7OADYCpwLX5z+kXwMThx5XqncinZzTE7Odhh7rwXb7488EcGT+4tWQeNceZ6pksy1/gU7M43x6Nx5nYBYvTTbFj+twfTR9fBltdAZ/oAcNZOyAk5cN3gTcAkyNiEdy0aPA1Jwebn+b4gMt4p32ReCTwG9z/lXAExHxfM7Xx/m7fcvle7L93v636KTZwE7gm3np8BuSDqOLj3NEbAf+CvgV8AjVcdtCdx/nQeNxXIfrY1hONoakw4HvA5+IiCfry6L6X5euqY+X9E5gR0Rs6fRYxtFEqkstKyPiTcDTVJc+fqcLj/MUYBFVov2XwGHAwo4OqgPG47i224eTzehsB2bW5mdk7IAh6eVUieaKiLg2w49JmpbLpwE7Mj7c/jbFZ7SId9JbgHdJ+iVwFdWltC8BkyUNvrm2Ps7f7VsuPxLYxd7/t+ikAWAgIm7J+Wuokk83H+ffBx6KiJ0R8RxwLdWx7+bjPGg8jutwfQzLyWZ0NgNzssJlEtWNxXUdHlPbsrLkMuC+iPh8bdE6YLAiZQnVvZzB+OKsapkP7MlT6fXAAklT8v8oF1Bdz34EeFLS/OxrcW1bHRERF0XEjIiYRXW8fhwR7wNuAt6dzYbu8+B/i3dn+8j42VnFNBuYQ3Uzdb/7mYiIR4Ftkl6XodOAe+ni40x1+Wy+pH+RYxrc5649zjXjcVyH62N4nbyJ1w0fqgqPf6SqTPlUp8ezl2N/K9Xp753AHfk5k+pa9UZgK/C/gaOyvYCv5L7eBfTWtvUBoD8/59bivcDduc6XGXKTusP7fwovVqMdS/VLpB/4HnBIxl+R8/25/Nja+p/K/bqfWvXV/vgzARwP9OWx/iFV1VFXH2fg08Avclzfoqoo66rjDFxJdU/qOaoz2PPG47gO10fTx4+rMTOz4nwZzczMinOyMTOz4pxszMysOCcbMzMrzsnGzMyKc7IxG2OSfpbfsyT94Rhv+89b9WW2v3Pps1khkk4B/jQi3rkX60yMF5/d1Wr5UxFx+FiMz2w8+czGbIxJeionLwX+raQ7VL1bZYKkz0nanO8T+VC2P0XS30taR/VX7kj6oaQtqt7HsjRjlwKH5vauqPeVfxX+OVXvbrlL0ntr275ZL77L5orBd5KYjaeJIzcxs320jNqZTSaNPRHxZkmHAD+V9LfZdh5wXEQ8lPMfiIjdkg4FNkv6fkQsk/SRiDi+RV//geopAW8Ejs51fpLL3gS8Afg/wE+pnhH2D2O/u2bD85mN2fhZQPVsqjuoXuXwKqpnbQHcWks0AB+T9HNgE9VDEufQ7K3AlRHxQkQ8Bvwd8Obatgci4rdUjySaNSZ7Y7YXfGZjNn4EfDQi1r8kWN3beXrI/O9TvczrGUk3Uz27a189W5t+Af+7tw7wmY1ZOf9E9brtQeuBP87XOiDpX+dLzIY6Eng8E83rqV7LO+i5wfWH+HvgvXlfqIfqdcG3jslemI0B/x+OWTl3Ai/k5bDLqd6bMwu4LW/S7wTOarHej4APS7qP6knDm2rLVgF3SrotqlcjDPoB1auKf071JO9PRsSjmazMOs6lz2ZmVpwvo5mZWXFONmZmVpyTjZmZFedkY2ZmxTnZmJlZcU42ZmZWnJONmZkV9/8BUcsxCSSvhSgAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Plotting the Test Errors\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}