raimonizard · December 9, 2024 10:19
diff --git a/machine_learning_intro.ipynb b/machine_learning_intro.ipynb
 {
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/raimonizard/60abb6bdf76c8b3157e9632b212f4977/machine_learning_intro.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# About"
      ],
      "metadata": {
        "id": "l1uWVK2uWpLS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "An exemple in order to work with the basics of **Machine Learning (ML)** with Python.\n"
      ],
      "metadata": {
        "id": "UtWQe9dgRrcd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Theory"
      ],
      "metadata": {
        "id": "BmKwwiIsl6mN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Machine Learning"
      ],
      "metadata": {
        "id": "Kqtnfo5B4c49"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Machine Learning** is making **the computer learn** from studying data and statistics.\n",
        "\n",
        "Machine Learning is a step into the direction of **artificial intelligence** (AI).\n",
        "\n",
        "Machine Learning is a program that analyses data and learns to **predict the outcome**."
      ],
      "metadata": {
        "id": "0SOGlkSA4hiN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data Set"
      ],
      "metadata": {
        "id": "4njstV_h5Ob7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In the mind of a computer, a data set is any collection of data. It can be anything from an array to a complete database."
      ],
      "metadata": {
        "id": "0MsgaooA5R8m"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Mean, Median and Mode"
      ],
      "metadata": {
        "id": "O2j0DD4y5XwP"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In Machine Learning (and in mathematics) there are often three values that interests us:\n",
        "\n",
        "- **Mean**: The average value.\n",
        "- **Median**: The mid point value after sorting the array. If there are two numbers in the middle, divide the sum of those numbers by two.\n",
        "- **Mode**: The most common value (the one which appears more times)."
      ],
      "metadata": {
        "id": "fLwheQwM5Z_M"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from scipy import stats as st\n",
        "import random\n",
        "\n",
        "# Use random.choices to select k elements within a range with the chance of repeating values\n",
        "# Used range(1, 220, 1) means values from min = 1 to max = 220 with 1 step between values.\n",
        "# So values in (1, 2, 3, 4, ..., 220)\n",
        "# [Source](https://pynative.com/python-random-sample/)\n",
        "speed2 = random.choices(range(1, 220, 1), k = 15)\n",
        "\n",
        "print(speed2)\n",
        "\n",
        "mean = np.mean(speed2)\n",
        "median = np.median(speed2)\n",
        "\n",
        "# Note that for mode() we use scipy.stats library instead of numpy:\n",
        "mode = st.mode(speed2)\n",
        "\n",
        "print('Mean: ', mean, ' Median: ', median, ' Mode: ', mode)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z8Wutgwc5pW3",
        "outputId": "98dada64-0ed8-4985-ac87-a21967d87450"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[169, 2, 7, 11, 73, 130, 68, 75, 8, 65, 51, 139, 145, 55, 194]\n",
            "Mean:  79.46666666666667  Median:  68.0  Mode:  ModeResult(mode=2, count=1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Standard Deviation σ"
      ],
      "metadata": {
        "id": "EbNk337K_kli"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Standard deviation** is a number that describes **how spread out the values are**.\n",
        "\n",
        "The standard deviation uses the same data unit as the values inside the dataset.\n",
        "\n",
        "- A **low standard deviation** means that **most of the numbers** are **close to** the **mean** (average) value.\n",
        "\n",
        "- A **high standard deviation** means that the **values are spread out** over a wider range."
      ],
      "metadata": {
        "id": "kK8DZtm9_oVA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "speed = random.choices(range(0, 220, 1), k = 400)\n",
        "print(pd.DataFrame({ 'speed' : speed }))\n",
        "\n",
        "mean = np.mean(speed)\n",
        "print(mean)\n",
        "\n",
        "standard_deviation = np.std(speed)\n",
        "print(standard_deviation)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6ZHj_UuQAMCp",
        "outputId": "ac385e11-b27d-4033-e14e-bd617b0488f0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "     speed\n",
            "0       13\n",
            "1       21\n",
            "2       36\n",
            "3      147\n",
            "4       37\n",
            "..     ...\n",
            "395     40\n",
            "396    149\n",
            "397    100\n",
            "398    188\n",
            "399    116\n",
            "\n",
            "[400 rows x 1 columns]\n",
            "107.84\n",
            "61.76649091538226\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Variance σ^2"
      ],
      "metadata": {
        "id": "2zoiwAHzCCH2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Variance** is **another number** that **indicates how spread out** the values are.\n",
        "\n",
        "In fact, if you take the square root of the variance, you get the standard deviation!\n",
        "\n",
        "Or the other way around, if you multiply the standard deviation by itself, you get the variance!"
      ],
      "metadata": {
        "id": "OG6BuunUCD4I"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "To calculate the variance you have to do as follows:\n",
        "\n",
        "1. Find the mean:\n",
        "\n",
        "(32+111+138+28+59+77+97) / 7 = 77.4\n",
        "\n",
        "2. For each value: find the difference from the mean:\n",
        "\n",
        " 32 - 77.4 = -45.4\n",
        "\n",
        " 111 - 77.4 =  33.6\n",
        "\n",
        " 138 - 77.4 =  60.6\n",
        "\n",
        " 28 - 77.4 = -49.4\n",
        "\n",
        " 59 - 77.4 = -18.4\n",
        "\n",
        " 77 - 77.4 = - 0.4\n",
        "\n",
        " 97 - 77.4 =  19.6\n",
        "\n",
        "3. For each difference: find the square value:\n",
        "\n",
        "  (-45.4)^2 = 2061.16\n",
        "  \n",
        "  (33.6)^2 = 1128.96\n",
        "  \n",
        "  (60.6)^2 = 3672.36\n",
        "  \n",
        "  (-49.4)^2 = 2440.36\n",
        "  \n",
        "  (-18.4)^2 =  338.56\n",
        "  \n",
        "  (- 0.4)^2 =    0.16\n",
        "  \n",
        "  (19.6)^2 =  384.16\n",
        "\n",
        "4. The variance is the average number of these squared differences:\n",
        "\n",
        "  (2061.16 + 1128.96 + 3672.36 + 2440.36 + 338.56 + 0.16 + 384.16) / 7 = 1432.2"
      ],
      "metadata": {
        "id": "Ko6JW-pDCdnr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy\n",
        "\n",
        "speed = [32,111,138,28,59,77,97]\n",
        "\n",
        "x = numpy.var(speed)\n",
        "\n",
        "print(x)"
      ],
      "metadata": {
        "id": "hEGW-kIdDCPf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8b5e98ed-7dff-4aa1-a73f-f03b83a16a86"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1432.2448979591834\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Percentiles"
      ],
      "metadata": {
        "id": "595_JdAkDp4n"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Percentiles are used in statistics to give you a number that describes the value that a given percent of the values are lower than."
      ],
      "metadata": {
        "id": "ECrBHph4DsAU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "ages = random.choices(range(0, 100, 1), k = 100)\n",
        "print(ages)\n",
        "\n",
        "# Sort the values of ages list ordered from small to bigger\n",
        "ages.sort()\n",
        "print(ages)\n",
        "\n",
        "# Find the age value from which the 50% of the population is equal or younger\n",
        "print(np.percentile(ages, 50))\n",
        "\n",
        "# Fint the the top 10% younger age:\n",
        "print(np.percentile(ages, 10))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eKmUHYveD193",
        "outputId": "496fd273-aac1-4270-b9c2-a633f984e36d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[51, 55, 3, 35, 78, 89, 41, 18, 23, 37, 75, 31, 29, 20, 98, 97, 11, 91, 92, 69, 16, 50, 77, 45, 23, 58, 58, 99, 49, 49, 55, 82, 98, 59, 38, 48, 39, 17, 47, 94, 78, 26, 57, 66, 39, 5, 9, 20, 98, 45, 89, 60, 2, 78, 11, 99, 87, 12, 21, 37, 71, 84, 22, 85, 75, 72, 28, 36, 25, 0, 45, 98, 90, 8, 25, 67, 46, 89, 29, 95, 6, 34, 78, 75, 60, 19, 99, 17, 2, 76, 31, 36, 65, 1, 71, 63, 11, 90, 52, 49]\n",
            "[0, 1, 2, 2, 3, 5, 6, 8, 9, 11, 11, 11, 12, 16, 17, 17, 18, 19, 20, 20, 21, 22, 23, 23, 25, 25, 26, 28, 29, 29, 31, 31, 34, 35, 36, 36, 37, 37, 38, 39, 39, 41, 45, 45, 45, 46, 47, 48, 49, 49, 49, 50, 51, 52, 55, 55, 57, 58, 58, 59, 60, 60, 63, 65, 66, 67, 69, 71, 71, 72, 75, 75, 75, 76, 77, 78, 78, 78, 78, 82, 84, 85, 87, 89, 89, 89, 90, 90, 91, 92, 94, 95, 97, 98, 98, 98, 98, 99, 99, 99]\n",
            "49.0\n",
            "11.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data Distribution: Histogram"
      ],
      "metadata": {
        "id": "HBzCsi8JG2l7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "To visualize the **data distribution** set we can draw a **histogram** with the data we collected.\n",
        "\n",
        "We will use the Python module **Matplotlib** to draw a histogram.\n",
        "\n",
        "The **histogram** graph is a **two axis bar diagram**.\n",
        "\n",
        "- The x-axis: contains ordered values form the dataset.\n",
        "- The y-axis: contains the count of elements inside the dataset which have its value inside each x-axis value.\n",
        "\n",
        "The histogram will have multiple vertical bars.\n",
        "\n",
        "It might be interesting to check out the standard deviation of the data in order to know how many bars to be used in the histogram.\n",
        "\n",
        "As more bars, more detailed information about the data distribution.\n"
      ],
      "metadata": {
        "id": "nokQI-r8G6_R"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Example 1 with 400 integer pseudorandom values between 0 and 1000:\n",
        "dataset = random.choices(range(0, 1000, 1), k = 400)\n",
        "dataset.sort()\n",
        "\n",
        "print(type(dataset))\n",
        "print(dataset)\n",
        "\n",
        "# matplotlib.pyplot.hist returns 3 arguments (counts, edges, bars)\n",
        "# Calling the function plt.hist(dataset, number_of_bars) generates the graph\n",
        "#  and we can also get the output arguments to be used later for the bar_label\n",
        "counts, edges, bars = plt.hist(dataset, 20)\n",
        "plt.bar_label(bars)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O6iw4QrTIcZh",
        "outputId": "b83832b4-46c6-45b4-8a23-8d8117966dec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'list'>\n",
            "[2, 6, 7, 8, 11, 13, 15, 17, 20, 23, 23, 24, 26, 28, 29, 32, 35, 36, 39, 45, 47, 52, 52, 54, 58, 60, 60, 65, 66, 69, 69, 72, 73, 78, 82, 82, 88, 90, 91, 92, 93, 93, 97, 99, 101, 107, 111, 111, 113, 114, 115, 119, 120, 126, 129, 131, 138, 140, 143, 143, 145, 145, 146, 149, 156, 161, 164, 165, 168, 169, 180, 180, 184, 185, 186, 188, 188, 190, 191, 198, 204, 207, 209, 216, 217, 218, 218, 223, 227, 231, 233, 234, 241, 244, 245, 246, 247, 250, 255, 256, 259, 264, 268, 268, 274, 280, 283, 288, 291, 297, 299, 300, 301, 301, 306, 312, 317, 317, 317, 322, 324, 327, 329, 339, 345, 348, 361, 361, 366, 366, 367, 372, 373, 375, 376, 377, 379, 379, 380, 383, 385, 388, 391, 392, 393, 398, 399, 400, 401, 401, 401, 403, 405, 406, 407, 408, 408, 408, 410, 411, 413, 414, 416, 421, 424, 425, 435, 438, 438, 439, 440, 441, 442, 450, 452, 459, 460, 462, 463, 471, 474, 477, 478, 482, 492, 492, 494, 495, 498, 498, 498, 501, 502, 507, 509, 516, 523, 524, 524, 527, 530, 531, 531, 531, 532, 533, 535, 540, 542, 542, 542, 544, 544, 547, 548, 555, 557, 558, 559, 559, 562, 567, 568, 568, 570, 572, 575, 580, 588, 593, 597, 597, 598, 598, 600, 602, 602, 603, 604, 607, 607, 608, 609, 616, 617, 618, 620, 626, 632, 635, 637, 638, 642, 648, 649, 650, 653, 655, 659, 661, 664, 666, 671, 673, 678, 681, 685, 691, 694, 697, 697, 701, 705, 707, 711, 711, 711, 713, 714, 714, 717, 718, 721, 721, 723, 725, 725, 726, 735, 738, 739, 741, 746, 747, 751, 751, 751, 752, 761, 762, 763, 765, 765, 766, 769, 773, 775, 778, 780, 780, 783, 787, 788, 791, 793, 793, 795, 798, 799, 802, 803, 803, 804, 805, 806, 814, 820, 823, 824, 825, 825, 828, 829, 829, 830, 831, 833, 835, 835, 836, 837, 839, 843, 845, 845, 847, 852, 852, 858, 860, 862, 865, 865, 872, 873, 873, 877, 879, 881, 884, 890, 891, 892, 892, 900, 900, 902, 902, 914, 916, 925, 931, 934, 935, 937, 940, 941, 942, 944, 944, 951, 951, 952, 959, 962, 966, 966, 972, 980, 981, 982, 985, 986, 986, 989, 991, 992, 996, 998, 998]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Example 2 with 400 float pseudorandom values from 0.0 to 100.0:\n",
        "dataset = np.random.uniform(0.0, 100.0, 400)\n",
        "dataset.sort()\n",
        "\n",
        "print(type(dataset))\n",
        "print(dataset)"
      ],
      "metadata": {
        "id": "ZNgm9g3dKqYY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "counts, edges, bars = plt.hist(dataset, 10)\n",
        "plt.bar_label(bars)\n",
        "\n",
        "plt.show()\n"
      ],
      "metadata": {
        "id": "1OmUVDC-L5Ae"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Normal Data Distribution"
      ],
      "metadata": {
        "id": "tYZ6owkJQhqZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this chapter we will learn how to **create an array** where the **values are concentrated** **around a given value**.\n",
        "\n",
        "In probability theory this kind of data distribution is known as the **normal data distribution**, or the **Gaussian data distribution**, after the mathematician Carl Friedrich Gauss who came up with the formula of this data distribution.\n",
        "\n",
        "Normal data Distribution is commongly known with the term **nd**.\n",
        "\n",
        "**Note**: A normal distribution graph is also known as the bell curve because of it's characteristic shape of a bell."
      ],
      "metadata": {
        "id": "SdSzFfc-Qj6l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# The instruction numpy.random.normal(center_value, standard_deviation, num elements)\n",
        "#   generates a numpy.ndarray list with values around the median value = 5 with the standard deviation of 1.\n",
        "#   Assuming that we created a normal distribution series, the 68% of the values of the ndarray will be between [4, 6]; because (5-1) = 4 and (5+1) = 6. Those values are called to be on one standard deviation\n",
        "#   The 95% of the values will be between [3, 7] within two standard deviations (standard deviation value * 2).\n",
        "#   The 99.73% of the values will be within three standard deviations values [2, 8].\n",
        "dataset_nd = np.random.normal(5.0, 1.0, 100000)\n",
        "print(type(dataset_nd))\n",
        "\n",
        "counts, edges, bars = plt.hist(dataset_nd, 20)\n",
        "plt.bar_label(bars)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "0e3U47glRCO9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Cars example with normal distribution"
      ],
      "metadata": {
        "id": "XJrOPcuIUztv"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Generate the two sets (cars age and cars speed)"
      ],
      "metadata": {
        "id": "AD3ztQOgaGGk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "cars_years_old = np.random.normal(5.0, 1.0, 1000)\n",
        "cars_speed = np.random.normal(50.0, 20.0, 1000)"
      ],
      "metadata": {
        "id": "H4hWXwBXaECS"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Plot the datasets:"
      ],
      "metadata": {
        "id": "lK9_MDZcaLP1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Generate a scatter plot representing all the points\n",
        "plt.scatter(cars_years_old, cars_speed)\n",
        "\n",
        "# Add a title for the graph\n",
        "plt.title('Distribution of speed and age for cars', fontsize = 14)\n",
        "\n",
        "# Add labels for axis\n",
        "plt.xlabel('car age (years)', fontsize = 8, color = 'blue', fontweight = 'bold')\n",
        "plt.ylabel('car speed (kms/h)', fontsize = 8, color = 'blue', fontweight = 'bold')\n",
        "\n",
        "# Show a background grid\n",
        "plt.grid()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "1MR16h7wU5Qs",
        "outputId": "3e11c175-9ba9-4dda-ea1c-293ea970f9cc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        }
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Check the mean and standard deviation of the generated sets:"
      ],
      "metadata": {
        "id": "KHBmU59vaNuv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "mean_age = np.mean(cars_years_old)\n",
        "print('The cars age mean is: ', mean_age)\n",
        "\n",
        "standard_deviation_age = np.std(cars_years_old)\n",
        "print('The standard deviation for cars age is: ', standard_deviation_age)\n",
        "\n",
        "mean_speed = np.mean(cars_speed)\n",
        "print('The mean value for cars speed is: ', mean_speed)\n",
        "\n",
        "standard_deviation_speed = np.std(cars_speed)\n",
        "print('The std deviation for cars speed is: ', standard_deviation_speed)"
      ],
      "metadata": {
        "id": "vCch_xZPZvWL",
        "outputId": "5fefbc1d-4c9b-4ddd-f6ed-0f4cd5c2b268",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The cars age mean is:  4.998381513061106\n",
            "The standard deviation for cars age is:  0.9892486206465505\n",
            "The mean value for cars speed is:  50.603779692796415\n",
            "The std deviation for cars speed is:  20.18948558606819\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## The mathematics line formula"
      ],
      "metadata": {
        "id": "4Bx1PIUTo5Mn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In mathematics, the line is represented as shown below:\n",
        "\n",
        "![image.png](data:image/png;base64,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)\n",
        "\n",
        "Source [here](https://www.geogebra.org/m/WCbjCGDC)"
      ],
      "metadata": {
        "id": "uB_ge5ITl999"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Regression"
      ],
      "metadata": {
        "id": "pB4-27EFo-dq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The **term regression** is used when **you try to find the relationship between variables**.\n",
        "\n",
        "In Machine Learning, and in statistical modeling, **that relationship is used to predict the outcome of future events**.\n",
        "\n",
        "There are **different kinds of regression** to be used depending on the distribution of the data."
      ],
      "metadata": {
        "id": "pGhSmSJao_8_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Linear Regression"
      ],
      "metadata": {
        "id": "01OHAB4ypE2w"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The **linear regression** is the **straight line** which represents the relation among two variables.\n",
        "\n",
        "Linear regression uses the relationship between the data-points to draw a straight line through all them.\n",
        "\n",
        "This line can be used to predict future values.\n"
      ],
      "metadata": {
        "id": "er1V0L5mpnMo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "6VnEqxyArYA0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Polynomial Regression"
      ],
      "metadata": {
        "id": "ZB7aadQ6qVdj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**If your data points** clearly **will not fit a linear regression** (a straight line through all data points), **it might be ideal** for **polynomial regression**.\n",
        "\n",
        "Polynomial regression, like linear regression, uses the relationship between the variables x and y to find the best way to draw a line through the data points."
      ],
      "metadata": {
        "id": "7SdEtGVTqX8J"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![image.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAssAAAIQCAYAAAB38SM3AAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAAFiUAABYlAUlSJPAAAKEoSURBVHhe7P0HfFzndS1uLwKDmUHvvRIsYG9ikURKVO+SVW3L3XEcJ7EdxYl9k9wk/yRXib/clOs4dhzHsh232Cq21SWqUKQoiUXsvYAkeu99AAzA76yNgUzRBAlSqDPr0W8EcOYMiAEH56zznv3ud8YZB0RERERE5LeEBT6KiIiIiMg5FJZFREREREagsCwiIiIiMgKFZRERERGRESgsi4iIiIiMQGFZRERERGQECssiIiIiIiNQWBYRERERGYHCsoiIiIjICBSWRURERERGoLAsIiIiIjKCGWccgc+DDl9ad3c3/H4/3G534F4RERERCVXMhRQTE4MZM2bY5xcStGGZL6unpwebN2/GsWPHEBERMaofiIiIiIgEJ+ZDhuWMjAzcfffdiIqKQljYhQstgjYsDw4OorGxEV/96lfxzDPPID093cKyArOIiIhIaGI+bG5uxsKFC/HEE08gJSXFBlQvJKhHljs7O/HII4+goaEB//iP/2hnDwrLIiIiIqGJVQff+ta3cOTIETz77LNWihEeHh549PyCOix3dXVZWGZo/u53v4vo6OiLDrWLiIiISHBiNvz617+Od99918Iys6HCshOWOcnvsccesx+IRpZFREREQhMz4aOPPort27ePOixrmFVEREREZAQKyyIiIiIiI1BYFhEREREZgcKyiIiIiMgIFJZFREREREagsCwiIiIiMgKFZRERERGRESgsi4iIiIiMQGFZRERERGQECstjjMsh9vkH0dM/YDd+HpxrJIqIiIhcnv6BQFbqG0Cvk5UGp3BYUlgeQ/x37nf+wU/Wd+LdkmbsON2E47UdzptgQIFZREREQh7j0MDgGZQ0dGFXaTO2O1npSHU7ununblaaccYR+PyC2tvbcerUKbzzzjsoLS1Fc3MzoqKi8NGPfhSLFi1CQkJCYEuGxjM4ePCgbbtv3z74fD6kpKRgxYoVuOGGG5CWlvbeOtyDg86ZRU8PXnnlFezevRvV1dUICwtDUVERrr/+eixZsgQej8e2vRT8Hrq6uvDII4/YOuCPPfaYrf89Y8aMwBZjq6atB/srWrHlRCOqWnvQ1ee3d0SkOxzZCZFYO9t5/XkJyE6MCjxDREREJHQ0dPTiUFUbNh2vR2VLDzp8fhtR9rjCLCutKUzCqoIk5CdHB54x9pgJH330UWzfvh3PPvusZcPhTDqS8L91BD6/oNbWVhw7dgw7duxAY2Mjjh49ij179mDlypXIy8tDTExMYEvgxIkTePHFF7FlyxYMDAxY+G1oaEBlZaWF7rlz51oAZnDl12VQfumll1BVVWXfMJ/Dr9HZ2WkvIjc3N/CVL01/fz82bNhgH++55x643e5xCct17T5sKW7Ec/uqsdl5A5xyzpaqnDdBdZvP3gylTV1o6e6zn0NqrAcxHte4hXYRERGRqYZBmaPIz+yrwsZj9Siu67SMVBPISmXN3Wjs7LVt0+M8iHKyUtg4ZCVmQuZTZtKHH37YsiHz2YWMugyDIZaBmCO9DzzwANatWxd45P04ovvaa69h8+bNFgi/8IUv4C/+4i9w1113Wfj94Q9/iJKSEhttZijmN/ud73zHgvLatWvxta99DV/5yleQmZmJt99+20J3b2+vfd2piN/VnrIWPL+/Gm84//jtzlkSLy8M4xlTp3MfR5yfc7bZWdIM/1mPi4iIiAQzph6OKL94oBobDtWitbv/fVmIn3X1+rHjdDOe2VuFd042Wi3zVDHqsJycnIxrrrkGn/nMZ3D33Xdjzpw5gUfej6F248aNVqLx0EMP2XNYUsFyDQZmjiS/+eabqKurs1Hm4uJiG62+/fbbcf/991tJBwP57//+7yMpKQl79+61IO33+wN/w9TBf9xB5x/7xYM12F/ZZn8+n+H7j9a0O2dU1VbIPkWzv4iIiMiYYlbigOKOkuaLZqWSxi48vrMCbT39UyYrjTosc5SYo8sRERFwuVznHbLu6+tDbW0t6uvrrUZ53rx59hxu6/V6kZ6ejpkzZ1pAbmtrQ1NTk40ysyRj9uzZyMjIsG359VmqwdFl1jMfP37cvvZIGNA5rM5tWYty9m08Q7bPOevhBD6rUXbOiC6E/97dzvbVzrYnnOdYTbOIiIhIEGNXsFP1naho6bYa5QthVvL1D1ppRnFdB9p9/UMPTLJRh+XRYGBlbTJDamxsrE3kG8awzTKO1NRUC9MMtizLYGCOi4tDfHy8Beph3Jb38Xk1NTVWsjESjlC/9dZb+Pa3v41/+Id/sNvXv/51/PM//zP2798f2Grs9Q3wH5RBeeB9pRcj4TZdTmDmm4CjyyIiIiLBjOUWnNvFoDyaMtSBM2espRyfw0HGqWBMwzIDLQMwP3K0+OzwSxwxjoyMfG/El+Ga9cgs2TjfTEQWXfN+drVg14yR8GuUl5dj165d1oFj+MaZjiz3GC8c0WZgvpR6ahsFv8TniIiIiExPTu5xMtyl9lHud4L1VOm9PKZhmTgSzNtIGBIv9Pj5XGx7lnxwRuMPfvADPPfcc+/dnnzySdx0002BrcaeKywMSVEeuMJH/2PkcxKj3Ii4hOeIiIiITEfsaJEQ6YbHdeH2bGcbek6EtZSbCsb0u2A9MyflcQSZo8cdHR2BR4aw7pj3sbyCo8YceWa5BSf9na+2mKUaHH1mD+cLtfXgY/x6HLVmqzneOFrNj/yexgt7KM/PjEVmvBdRzucXivR8LDIiHOnxHsxznhPtcQ09ICIiIhKk3E7gnZMei6wEr2Wfiw2XMiCnxXow13lOvBOYp4IxD8ucpMdw29LSYqURw1hGwfsqKiqsbzJDMrfLzs620g0uRsJJf8TRZ9Yps/6ZwbuwsPCioZejzyzZOPvGEH2po9iXIjxsBmK9EbhmTgpmp/2mz/S5hr+DgpQoXF+UhoQoN1zOc0VERGTqYzUA28CebujC4eo2W6m3qbPP7g8GfB2sD2YnCq6mx8l19R29Y1IGwVFiri9xZWEyFmbF2X0jJSDen5UQiVsWpNu6FFPlKvyovwurte3vt8l0HAnmqC8DMEeQeR9HjDk6zFFjtorjNtu2bbMQzM/Z0eLQoUO27bJly6wVXWJiogVhjkZzgRNOxuPKgFz0hO3lOBGQHTG4DUeOpyJm8XVOWL6uKBVznMDsdv5hz87n/JxnSbNSY3DtnFQnLKc6//gjvU1ERERkqmBWbOrsxb6KFrxyuNYW1Pj1nio8vbcKLx2swdZTjdblajpP2m/u6sPBqla86ry+Z53X9eu9lfb6XjxQg7eKG1HR3A1f/28m2jEP8s/sVMHnciIeu4KVNXU5JxOdKHZOJHgyccr5nCcXDOBlzV3IivdiaU488pOj4IkI+60FR5iVCpzH1s1Owa2LMixgTxWjXsGPQZmh9/Dhw7aQyM6dO20pa7aC4+gtJ9kRAzAXHGEw5rYMzwzSXEmPk+7Y+eKLX/yijUCzDIPPZfs4Lo/N4M1QXFZWhh/96EcWvtevX48bb7zxgmUYI+H3PBEr+CVHe5Ae50WkJxzlzhtiuDsGz4jivBHISYzEfSuyceeSLBRlxI3L9yAiIiJjh6OqXBhj2+km/HRbGX7+bjm2nGjA3vJW7Cxtxo6SJhtlZmlBaszQ6rzjseLceOGJAJsU7Cxtwi/ercCPt5Vis/P69tjra7HXt7e8xUpOeRXdFT7DRp/be/wWjsubui0IMxgfq+1wfhbtOFjZ6gTvNhyraXfu67RWuRylZoBmNwyWZPDnVN3WYy3l+D3wSnuk22VZ6Y4lmbh3eTYWZsXb1fvxwEx4qSv4zXDOEEY1xs4Wb+wu8Td/8zc24svSCY4Ss5SCnS8WLlxogfRLX/qS3c8lrLnmNkeLGaQZoq+88ko8+OCDtkT2cFBmsC4tLbVwzJFodq/gN83RZG57yy23ICsrK/BdjB5fFrtoPPLIIxbCH3vsMathHq+gyg4X7T399sb5X788YB8XZcfjY6vzcEVBov0i8ReKbxQRERGZ2jhqussJxf+x6SRONnRZcD67LIFZjleTGSR/b30hbl6QjoLk6MCjU1+Hrx/7Klrxn5tPOSG3Dd3972+Dy7jEpgRxkS67cs76YQ4G1rf3wucfsDZwXGyEPxM+jR+ZvYa/wozhYgvnAz/jz4s/PgZ0hm4+l4OK2U5IvnZOCu5Zlm0/vxiva1wn9jETPvroo5ZpmVOZDVm6eyGjDssMtQyy7777rv1F52IY5ijz4sWL7c8sv+CIMVffY4rnN8NaZa78x3rl4RTPv56PHzt2zOqZWbfMQMsezdyWQZl1y5dqosMy8Y3CwPxHj+/FnrJWm/z3++tn4epZKXZmJiIiIlMfkxGXZ/7mxhO26tyFFtNgqrgiPxGfvCof9yzNtpA51fH1lTR14V9fOY7tp5vQ5JwYXAhHg1lC2j8w1AOZr9HrCrfmBt6I4VuYE3LDbQSaGWw4QNtH5398Hhdw6wzcGMyZyeY6QfyzV8/ETQvTkRw9/iW34xqWp5vJCMvEM89/eOmILeuYHOPBp51fntsXZyLare4XIiIi0wHD8abj9firZw5ZwLvQwmNMFVGecHzqqgJ8/ppCJE1A4Psg+FoaO3ux9VQTHn3hiA3yjWaxEAbZmSnRVnYa63XZ1fIo5+Z1hb0XljnSzna6w2F5+CMXGmGdc7cF5QELy/wzb/x6dzo5aV5mnIXv8aawfJbJCsss8v/ptlKbANDd57egzNHlqdL+RERERC6stKkLL+yvwb+8ejxwz8Xduywbn1s3E4tz4gP3TB0Mw702sjuANl+flV28fqQeGw7X2ujvaKzIS8SHV+Zg7ewUKy31OAH5ciMVMxpLMdqcoM4yVp5gRLld41anfLbLCcsqoB1j/HcucM68EqIi0O6cmZY0dMI/EJTnIyIiIkGJV4k5+nkpOEra0dsf+NPUwWDa2tWHd1l/vfkk/vjxffg/LxzB68fqRh2UKdIdZu3c0uK88HBtiQ+QazlwyXCcEe9FTmIUYjwRExKUL5fC8hjjTFi2PmEvZf6ilTV3/9akABEREZm6WFZwqSUBnMDPADgVsNSCrd22nWrCd988ZeH4314vxmtH6nC6scsJ9oNWMnEpWI8cFzkUasci1jJsMzPZ15u6OdkoLI8x/sOz/iYpOsJGmVu7+1Hb7rPALCIiIlMfrw5zcYxoT7gd1y+G4ZqrzmU4x//JxNFttnJ79UgtfvROqbW7e2ZfNTYfb7AWbjHO61k3KwUfXZWLe5dlIc47utIH1iWnxHiQ7fxMpvAA8LhRWB5j/J3irFG+qXgGxj6CJU2dl3w5R0RERCZHQuTQZLZ5GbE2ce1C+ZCP5SdF2Uq+aXGeoTsnEEeRmTHKm7uxp6wFLx6ssaD8nc0nbSSZLfAy471YPTMJdy3JwmfXFuAPr59lExLZzziaJRWBr3U+fIzt3dg+jmUT06mX9FhRWB4HrMXh2WVWfCT6nTfxqfouq18WERGRqY95MM8JwJ9Yk4/EqKHuFueLiMP33bow3drHTUaQ5ES5ozXt+N6bp/HVXx7Av28stnZ3bPOW7oT32xZm4M/vmIdvfWw5vnTDbFw9OwVpsV6kO8H3M2vznSAdaV/nQq9v/dxU53nJIRmUSWF5nLAUg2dy/oFBW/6Rzb9FRERkemApBjs//MF1XC8h2VqlnY2lFxxN/qMb5+C2RRnIccL1ROFocmtPH57bX41/ePEI/u65w1Z6wZZwbPF2w7w0/NWdC/DPDy7B714z0zpZREW8f4VBvp7VM5PtcW7PdnBn48Iguc5r+sL6Qty9NMtG2kOVWseNEy77+Ks9lbZEJtdB/6s75+O6orTAoyIiIjLVcXJ+dWsPjjjHdC7pXNXSY4trcDIfO0PMSo3B0px46+gwEYuPMSTXtftsme3dZS04WNVuNcqczJcW48Ei53tZkBmH+c6tKD3WykK4St6F1Div76jz2jg6XcGmBM7r48p9yTFDpSjLchMsNLPENBioz/JZJjssN3T04um9Vfi/G47Z2dnX71uMOxZnarlrERGRaYZJiT2BGZxZHxwZER5oo+aZkNIEhmTWHjPM7q9sxTsnG7HtdBPYl2I41C7KirdSCYZkLop2Kfj6+LoqW7rtI7NKSrQHmQnBV6OsPstTCBtsc2YsW7PwLK3eCc+8ZCIiIiLTC/MiyzIWZMXZRDkuPDIRk90YYtkooKmz1wLyNzcWWwu4TccbMDg4tK7D/ctz8Be3z8dXby3C1bNSLjkoE18GyzA4Ir2qIAlLcxJsUl+o1iifS2F5nISFzbBeyyzB4OWM6rYe1LX3Bh4VERERubDW7j68eaIef/f8EXzj9RNWesFRZnamYEeLv7l7AT62Js9Cs4wfheVxwnOxuEgXClNi4AqfgZpWH+rbfUMPioiIiJwHy0h7/QPYfroJP9paiv9+pxQ7SpqsVnmWE5I/sirXJhXeuTjTWr+xHITlnjJ+9NMdR3HeCBSmRtvIMgvoWYohIiIicj5sA8cV9l49XGdNAp7ZV4X9Fa2IdrusScB9y7PxwBU5uGVBupMvYi55lUG5PArL42goLMcggiPLzhkhzwq17LWIiIicbXiCHVfZe+lADf5pwzE8v78aTZ19VmJxsxOO//SWIlt5j90uXJe4VLV8MPppj6NYK8NgzfIMm8XKDhlc9lpxWURERIhlF/0DA9hwqMbqkh97q8QG2DyucNy+KBNfdUIyFxMpSI6y+2TiKSyPI9YQJcV4rJ6IfQ5buvtR2dJjvxgiIiIS2jiavL+yDf/8ynH8fEcF9lW0IsxJZlcXpuBPbp6LT1yZh+V5CdYwgDlCzSkmh8LyOGLLFTYpz0uKtrXXOau1rKkLg8rKIiIiIYtjZuXN3dh8vB5P7CzH8/trcKKuA+mxXtyxKNMm8XFthoXZ8RaUZXIpLI+zCOcUkZP8Yr0RTljuR2lTNwaVlkVEREIOQ7Kvf8AJyl3YdKwe/7OjHL/eU4WuPr/VJt+5JBOfvrrAPvKqNMs4ZfIpLI8zTu7jcphs9t3y3siywrKIiEio6RsYtBHlb7xWjP988xR2lDTDExGOFXmJ+LPb5uEzTlCekx4T2FqmCoXlccYao1mp0YiLjLClMsuauuF3flmUl0VEREIH11p4+WAN/vGlY9h6qtEyweKsOHz+mpm2+t6SnHhEe1xaNW8KUlgeZ+FhM5CbGIV4Jyz7B4bWdm/o7LWzSxEREQlu/c7x/lhNO57dX229k3eVNVsGuHZOCj68Ktc6XixyQjNzAjODTD0Ky+OMb/zE6AgkR7ttsh/rkkoau62FnIiIiAQndr5q7+nHkep2vHiwBk/tqrDlqnmlee2sFHx0VZ6twjc7LQYzNJo8pSksTwD+EmQlRCItzoPe/kEcr+1AV68/8KiIiIgEE1Za+vyDOFTdhm9vOokfvlOKk/VdTg7w2up7f333fFwzN0WdLqYJheUJkp3gRYbzS+LzD9gKPRxhFhERkeDT1t2Hp/dU4ttvnMS7Jc3oc47918xJwReuLcTn1s1EcrQH4TMUwaYL/UtNkOyESKQzLHNk2QnLnT6FZRERkWDS6x9ERXM3fvFuBZ7ZV40jNe3W/u3e5dn48MpcXDsnFdmJUVpgZJpRWJ4gmYGwzE4YbBvT2tPnnGlqkp+IiEgw6PD5cay2Hc8FJvIddYJyUrQbNy9It0VGrp2b4gTlyMDWMp0oLE8QNhdPY4Px8Bn2C1Xf3ouO3v7AoyIiIjJdcaERzkd6dm81/m1jsQ2K8Zh/0/x0fOXmuViWm2iLk8n0pLA8QVzhYUiOcSMnMcp6KPIXqaGjN/CoiIiITEfsesHV+H7w9mkbUR4YOIP5mXH47NqZzq3AOfZ71BJumlNYniD8NUmMcttylvydYU1TY0ff0IMiIiIyrTAkN3X24qndlXbbWdpig2E3zE/D7zgh+bqiVKTFeq1mWVF5elNYnkCJ0W7MTIlCmPOLU9Hco5FlERGRaWhg8IxdIX7taB1+6QRl9k/2RIRh/dxUfHhljn3MTojSiHKQUFieQIlREZiZHGNnnlWtTlh2zkj5CyciIiLTA4/bte0+bDregJ9sLcMeJyi7XWHW6eJja/Jwy4IM65+sbhfBQ2F5Ag2VYbBmGWjistcdvejU4iQiIiLTwuCZM2jr6cPPd5ThJ9tKUVzfaXOSPro6F5+8Kh/L8xICW0owUVieQDzzZGDman5u55eLtU5VLd2BR0VERGSqYseLU044ZreLVw/X2YBXXlIkvnLTHNyxKBMFydHWP1mCj/5VJxDLL6I9LsxKiUaUOxyNnX1W8yQiIiJTV1evH4er2/Hkrkq8fqQO9U5QnpMWiwdW5OD2xZkoTI2x47oEJ4XlCeZxhWFuRqwTmhmWe1HW1BN4RERERKYSzirqCfRQfvlQjROWK9Da3Y9ZqdG4c0mmrcqXmxhlx3YJXvrXnWDeiHDMs7DsQmNXL8qbuwKPiIiIyFQyOHjGWr3+bEcZfr6jHN19A8iKj7Qeyh9xgnJKrEcT+UKAwvIEY1geGll22dkpu2KwDoqTBkRERGRq4AT8A5Vt+LfXi/HOySarR16Wm4D/dVsR1sxMQqTKLkKGwvIEiwifgUznrJTrxbOGuc0JzJUtPejzDwa2EBERkcnU3NWHXaXN+NHWEuwoacLA4CBWFiTiE1fmYbUTlLUqX2hRWJ5gDMgxHhcy4ryI87rQ1TdUC+XrV1gWERGZbG09/dhX0YrnD9Tg5UO1VoqxPC8Rdy3Jwm2LMhEf6VZQDjEKy5MkNykSqbEem2F7zAnLnEAgIiIik4PVkP6BMzhc3War8j23rxrhM2ZgcU48Hl6d54TlTJvIpxrl0KOwPElyk6KQEuOxyQLHatqtbllEREQmR1efH3sqWvDYlhJsPdmIWK8LV89Kxhevm42V+YlwhSkyhSr9y0+SPCcsc2S52/nlPFHfYSPMZzTJT0REZMK1dPVhZ2kzvrflNPZXtsITEW5B+TNrC7AgK84JzhEaUQ5hCsuThJP8UmM8GHTyMZubc/lr1S2LiIhMLHam2l3eYmUXbx5vsHpkdru4e2mWE5hTEKOgHPIUlidJYlQEkmPctjgJO2FUNnejtacv8KiIiIiMJ17MZQnkoeo2PL+/Gi8erIErfAZW5CXg3mVZuHF+ugVn5WRRWJ4kM5zTVJZh5CdF2Z9Lm7rR3NVvn4uIiMj48vkHbAnr/36nBBuP1ds6CKvyk/A7a2fiylkpcKnjhQQoLE8iTvDLS46ys9vSpi7r6ygiIiLjq93Xj/0Vrfjum6fsY5Q7HGtnJ+MPr5+FeZlxiHSCs8gwheVJxKbmeUnROOP8V9LYhabOXgvOIiIiMj46fX4crGzD03uqsO1Uk13pXV2QhHuXZWNlfhLiVKMs51BYnkQpMW5rIceKqOq2HgvLfQOa5CciIjLWOBbV6x/E8boObDhUi+cPVNt9XML6ziWZVqPMmmUFZTmXwvIkSoxyIycx0pqc+/oGUe+EZQZmERERGVtcia+mrQe/eLccz+6rsiu5c9Ni8Mkr83FDUZpqlGVECsuTKMz5xeSS13PSY+CNCENdWy8qWroDj4qIiMhY6O0fQGljF/7jjZNWesHj79LceDxy0xwsyo6H26UaZRmZwvIk4jlsjMeF+RmxNpmgtt2HiuaeoQdFRETkA2N71hP1nXh8ZznePtloi4EtcQLyh1fm4or8RCREqUZZLkxheZJFMSxz5q07HHVOWGa/ZREREfngBgaHJtC/cbQev95bhbaefsxNj8VtizLsFuOJQJiSslyEwvIki3Y7YTlrqE0NwzLLMFhXJSIiIpePNckdPj9eOliDJ3ZWWFBOj/Xgw6tybXW+KOf4q5wso6GwPMki3WGYlRqDaI8L/QOD1muZExD86oohIiJyWQadpNzT78ePtpbitSN1FpSzEyLx5Rvn4KrCZAvKIqOlsDzJwsPCEOt12S8xA3Nbjx8nG7rQN6DRZRERkUt1xgnKdW0+/HJ3Jd44VofKlh7kJ0fh42vycaUTlLl6LpexFhktheVJxl9XlxOYC1KikRTltlWFjtd2oNc/MLSBiIiIjFpdey+2n27CL3dV4lRDF9LjPLhhXpqVXqTFeRERrugjl0bvmCmANVOFTljmin4dTlg+Uddhs3dFRERk9Hr6BrCz1AnKeypxsKoN7vAZuHlhOu5fkYPMeK96KctlUVieAriC38xUJyxHc2TZHxhZVlgWERG5FFtONOD5/TXYU95qq/E9vDoPdyzKRG5SZGALkUunsDwVOCe6eUlRSIpxwz9wBg2dvWjo8KGnX6UYIiIiF9PV58eu0mZbwnqvE5RjPS4bTb5pQbqVOar0Qj4IvXumAF4USohyIy3WgxivC919Ayhr6raWNyIiIjIyHjNP1nXiV3sqsbO0xQ6qq2cm4UEnLBelx9riXyIfhMLyFME6qvQ4L7LivRgYOIOT9V1o7e4LPCoiIiLnYpvVsqYubDregGf2Vtu8n6U5CbhzSSZWOYGZXaZEPiiF5SkkwwnLuUlR8A8O4mRDhxOW+wOPiIiIyLlYtrjxaB1+uq0UPv8AFmbF447FGbihKC2whcgHp7A8hWTEe5GXyLDMkeVOjSyLiIicBxcdae/pxzN7q/DyoVr4+gcwOzUGD63MwVWzkuGJCA9sKfLBKSxPIaxZzk4cmrHLPpFNnX3WBkdERESGMCi3dffbMtYbj9bboiNZzrHzoSucoFyYbCWNImNJYXkKiY90WylGjCfcQnJdh8+WvxYREZEhXLp6f2UrHt9ZgRN1nUiKduPaOal4cGUushIiEcbFC0TGkMLyFMKekAlREZiZHAO3KwxVrT2odm4iIiIytJT10ep2PPZWCY7WtNufr5mTgs9fU4jEKLeWsZZxobA8xcRFRqAoIxbu8DBUtygsi4iIDNtRMtRLmavznXH+/KHl2bhnaZaNLmtAWcaLwvIUE+sNhGWXE5Zbfahu8wUeERERCU0Dg2dwoq4Drx+twzunmqxl3PVFabhxfhrmBo6ZIuNF764pJtbrwpz0GHgiwqwlTq0Tlnv7B+wMWkREJNT0O8G40TkevnakDm8XN6K5sw/zMuLwwIpsLMlOQJw3IrClyPhQWJ5iuNJQYWo0ot3htty1LX3t3Aads2oREZFQcsY59HHNgd1lLXhyVwVONXTaMfK+5VlYOzsFyTHuwJYi40dheYoJC5uBKLcL+cnRNsrc0t1nPZd5CUpERCSUdPb6LSj/+8Zi1LX7UJASbaUXdy7JQqRbvZRlYigsTzGcnxARHobZadFIiIywsFzshGUuVCIiIhIq+gYGsf10E57ZV4XSpi7nmOjGzfPTnVuGdb5QiziZKArLU5ArbAbmpMXazqClqx/FdR22BLaIiEgoYPnF4ao2W8p6Z2kz3OHhuHZuKq6fNzShTzlZJpLC8hTEsDw33QnL0W5bzvNUfRd8fQO2apGIiEgw8w+cQWt3H57ZW413Tjaizz+IoowYfGx1HhZnx9sxUmQiKSxPQeFhYZiZEm0jyzYLuGuoK4avX6PLIiIS3HjMe2p3Jd462YC69l4bPPqjG+dgVloMPC7VKcvEU1iegnh5iRMXMhO8SIn1WEg+XteBDl9/YAsREZHgU++E4+2nmpywXIGaVh8WZcfj7qVZWJyTgCjnuKjyC5kMCstTFCcuZCdEIivei17/AI7WdKC9xx94VEREJLiw88We8ha8eLAGxXWdSI/zYn1RKq6dk2oT3rWUtUwWheUpLCcxEtmJUVavday2He0aWRYRkSDD2Thsj3q8tgMbjw0tPMI1B9bPTbUb+yqLTCaF5SksJ8EJy86t18JyB9q6+3FGk/xERCSIcNGtrl6/LTryxtF6uMJnYFF2HO6/IhuLsuICW4lMHoXlKSwtzovMeK/N/O3w+VHb7kNrj0aXRUQkOHD8p6mzD4+9ddoWH+kfHMQCJyA/cuNczEyORni4YopMPr0LpzBO8kuN9SDDCcxsG1fR3G2TH0RERIJBXYfP2sO9drTeuj5xjYG7lmRhWW4CYrwuW6hLZLIpLE9hnOSXFO3G7NQYm9hQ5oRl7kyGsSDD1z9gq/w1dPRaX0qWbKhQQ0REpjpO6DtY2YYXDlTjdH0nkmPcVqPMVfo4WHShFfpY48x5PE2dvWjq6rOrr6pSlPEy7mGZNbZ+vx+9vb3w+Xx24+e879z628HBQfT399s2PT099pF/DuU6XfZanpsRY6UY5QzL7T4Lw6zx6nWCckVzD3aVtmBLcQN2l7WiprUHff2DWsBERESmLB6iTjoBedOxerzh3AacO25flImbF6Tb1dSR8NjG9QcYlA9XtWPrqSZrNcdJ8F19fntMhz8ZazOcIDpub6uBgQG0tbXh6aefxvbt21FSUoK+vj7k5+fj/vvvx1VXXYWMjAzblt/G7t278fLLL+Ott95Cd3c30tLSsHbtWjz44IPIyspCRESEbTsa/HpdXV145JFH7Gs99thjiI6Oxoxp1qSRI8Zvn2zE3zx72DnLBj6ztgCfvboAR2ra8dz+audjh40oMzxzMROORC/PS8AdizOwKDsBHpcuHoiIyNTBUWEe27635RReOlhrIXftrBR81jm+Lc1NgDdi5IVHTjV0WreMV4/Uoa7dZ92iKNJ5Tnq8x0o4ripMRm5SlN0vci5mwkcffdRy6bPPPmvZMDz8wovdhP+tI/D5mKuoqMCvfvUrvPHGG/bNzJs3z4IyR4yPHDlio8Zz5syB2+3G3r178cwzz2DXrl2YO3eu3c/R59LSUpw8eRKLFy9GVFQUwpxAOFr8+hs2bLCP99xzj/090y0ss/yi39kZvHa0zjmT9ttIM09vnthZgZ2lLShr6kZzV589xsl//JwrHnFHxHrnOG8EPBfY8YiIiEwUjgy3OccqLjqy8ViDDfYUZcTh01fn2wIksc4x63wYsEsbu/DC/mrrw3y4ut2Oc/xavLV099uf69p6wULn1Bg34p3jpci5mAm3bNmCyspKPPzww5YNL5Ytx23YkSUV5eXleOKJJ+zza665Bp/+9Kfxuc99DnfddRfq6+vx9ttv48SJE/b4pk2bsG/fPiQlJeHjH/84fvd3fxf33Xcf4uPjLfkzXHd0dAS+eujg0p7JMUOT/CKc4HyirsPquzYdr0dNm88uOZ2NNcucCMjLWhucM/aq1h7VMIuIyJTQ0tWHnaXNeOlADapaupGXHIVbFqRjTWEyEi4Qbnmse6uYEwHrcKCyDT19A+8rt2AI73bu46Imrx2pw46SZpVjyJgZt7DMEoyGhgYcPHgQN998s4XlWbNmoaCgAB/60IewaNEiNDY2WrpnDfPWrVst3d97771YvXo1CgsLcfvtt+O2226z0g0+XltbG/jqv41lFyPdpjMOhEdGhGFeRixiIyNQ0tiN7aebnR1DYIPz4ENdzk7jeSdUsybMf06gFhERmWhDC2x14Gfby3C6sQtRbheunpWMh1bmIPoCS1lzVJkT+F4+VINT9V2Be3/b8GGRo84cMOLqt5q/I2NhXEeWGZgZVl0u12/Vg/C+1tZWHD16FNXV1RacOapcVFQU2AL2nJSUFLuP5Rjc/nz4d3BYnXUorFM++8ZSjunO7QrH/Mw4K6lg8PU5O4CL4c6Fl6VKnB3S2R00REREJgPn2nDUd1dZiwVnzq1hjXFcZMQFSyTZNYMhmyWGFzv+MRpz1JlXXo/WdMDXr8Ei+eDGLSwzDCckJCA7OxvvvPOOhWKGV3a52Llzp5VVsEyjrq4OTU1NNrrMmmQG5mH85fF6vXYfgzKfez7Nzc147rnn8Gd/9mf4/Oc//97tS1/6kpV6THeeiDDMz4hDfKTLdgSjPVFmYGYtF3c0IiIik4HHrOrWHut88eaJBvvzjfPTcO3cVMxMib5gizhisG7s7B1qjTqK4x9Hk9lWlc85t1RR5HKMW1jmqHBubi7uvvtu1NTU2OS9733ve/jBD35gn7NLBssuhtvEcRSa3S5439n4dTwej4VpbnM+DNV8bmRk5Hs3Bm9+ZGif7tzhYShIiXLCstsm/F2KS9taRERk7AzVEvux+XiDdbHgJLxZqdE2orwwKx7RntEdoy/nWKbjn4yVcQvLxM4XnNDHThaHDh3C97//fQvLx48fR2ZmJmbOnGnhmIGWgZelG7ydjX9mKQW3GWm2YmJiotU3/93f/R3+/d///b3bv/zLv+DKK68MbDV9MSCnxHicmxvR7tHvWNibmRMmRppdLCIiMp446Y4T03+9txLHnY+Z8V7ctCAda2enWMem0WAL1DRnW6/z8WKj0MRtuKhJWqwXEVouW8bAuL6LWELBSX3sTvfUU0/hpZdeso8MzMuXL7eRX/ZZTk5OtpFhjjC3t7cHnj2EtcjsghEbG/tbo87DGLT5GL8eW9TxxpFlfryU3sxTGV8jL1dlJYzcrP1sDNhsM8fnXKjBu4iIyHhgyURZUxf+Y9NJFNd1WknhyoIkfOqqAsRHjv7YzGWvizLjkB7nhdf5GheKy3wsygnKmfGRmJcZa/2XRT6ocQ3LDHgso2BwZSDmwiI5OTl2HxcoYVnFFVdcYffz8ZaWFhQXFweePTSqzHpk3peXl2c10BfCv+/sW7DJT46ys/JhI73C4Z3FvcuzMDttaKlsERGRiXS8th2vHK7DnvJWqzvmUtb3LM1CghOUL+W4xJFidsu4a2km5jjHNDrfs4fvW5wdh5vnp9mochBGAZkE4xqW2Z3i9OnTdmMbOYbhU6dO4fHHH7f7GJyvvvpqxMTEYOnSpTayvHHjRuu9zDZxnAi4bds2q1desWKFregXyvKTOEocaZ/z959n2OdeYuLlqoKUaNy2KMNuHInWvkJERCZSfUcvtp1uwsajdWjt7reVZa8rSsNCJ8hezgCO2zm2rZudglsXZmCF87U4IHR2ELZA7XFhTWESbnWOfatmJikoy5gZ17Dc2dmJw4cP4/XXX8crr7xitxdeeMFKMbjYyLp167BgwQKrRb7uuuusjpkr+bGzBZe95up/rHVmyQZHoM/ulBGKMhMikRHntZ0GdwIMxQsyYzE3PRazUmNQ5Hy8Ij8Rtzs7io+tznPOrhNUrywiIhOGpRfsQLGnrMU6XxTXd1op4F1LMnFFXqK1QL0cDMP5ydG4fXEmHrgiB1fNSnaOf3F29ZSjzYucEH717GR8ZGUubpqfjpxELXctY2fGmXFctYP9k1977TX813/9l03q40Q9jg4zGH/mM5+xAMy6ZuK38eabb+J//ud/LFCzTpkjz5y494UvfMEWKRmpZvl8+PXYqu6RRx6xEe7HHnvMapine3nG4zsr8G+vnUBtuw9/estcLHECMbX5+q1GebYTmrk+/mgmQYiIiIwlBuXmrj78718ftJFlhuNbFmbgC+sLkZ0wdGX0g2Jq6ekfwNGadmsPx+Md65nnZ8bCFabSC7kwZsJHH30U27dvtxWimQ1ZHnwh4xqWufIe+yOzlzJLLPhXccIdR5UZmvkNnt3hgu3kuDgJyzVYz8xwzDplTgJkqL6UoBusYXnj0Xr89zsleOdUI+5blo17l2djWW4C/INn4AqfAa8r3EaeRUREJhKPuxUtPfjellPYdKzBAu2amck2sJObFGVlgmNleHnroRVqZyDCNQNREeysNfS4yEguJyyPa6pi2GUoZus4LmG9Zs0aqz1mhwx2tzi3FRxDNB9buXKlbcvyC7aXY5eLYJywdzlSY92YnR7j7Bpm4FRjly06wtWPkqLddgavoCwiIpOhqrUHbxU3YMuJRrR292FJdjzuXJxhJYNjfWziaHKMx2XtUROiIqytqmKCjBclq2mGfSlZasGdQkVzN+rafejTCkUiIjKJOn1+7K9sw4ZDtahs6UF2YqSt0Md+yuz5rxwr05nC8jSTHO2x3skR4TPQ4py5c618zjQWERGZDCy/ONnQiS0nGqxEkCWBdyzOxDVzU5AYPfq5RiJTlcLyNMNWcZzIx+4XbLZe2+ZDSUNX4FEREZGJMzh4xgZsntlbhTePNyDWE4Eb5qVh/dw0a3cqEgwUlqcZll+wTmtR9tCa+uyKUdKksCwiIhOL3QHafX78ek8ldpY224S+OekxeHh1HgpTx75OWWSy6J08DUV5wm3iBEOzjSw3dlkrHRERkYnS1t2PA5WtePFgrXXByEmMtB7Hq2cmIc7rCmwlMv0pLE9DLL/gyDLDMntMljd1odc/YHVjIiIi4439lE81dOL5A9U4VtuOiLAZtlAI25myhak6WEkwUViehiLd4bZiEdfXHxg8g6bOPpQ3d9va+yIiIuONVzW3nmrESwdqrPzixvnpuNm5pcZ41MJNgo7C8jTE/pJeJzDnJ0fZTON2Xz+OVLfDp7AsIiLjiNcve/oG8Maxerx2pA4DZ85geW4CritKxbzMOISHKSlL8FFYnqYYmAtTY5AW60GHz49D1W22AxMRERkv/f5B7ChpwpbiBpQ2ddsy03cvzcKSnHjER0YEthIJLgrL09is1GikxTEs9+OwE5a7+wfsrF9ERGSssdSvpq0HT++twsGqNkRFhGNlQRJuX5SJzPjIwFYiwUdheRqblRaD9FgvOnsHcLy2E51OaGbPSxERkbFW39GLVw7XYfvpZjR39mFRdhw+t3YmkmPcKr+QoKawPI2lxHhsZDnGEw5f/wBON3TZqn4iIiJjiceWPeUteHJXhX2+qiAJtyzIQEFKNFxhihIS3PQOn8a8EeHIiPMiKyHSumIU13daZwwREZGx4neOL/vKW/Hq4VqUsU451oNr56ZgTWESotxsExfYUCRIKSxPcxnxXjuz54zk4roONHUpLIuIyNipaO7GW8UN2H66Cd6IMKybnYKrZqUgP1nLWUtoUFie5hiWZzpheXAQOFHfaYuUqGpZREQ+KB5LeNXy5UM12OoEZc6PyU6IxEfX5GF+ZtzQRiIhQGF5mmMZhoXlM2dQ3dqD+nYfunv9gUdFREQuDzstcTR549F6lDZ2ozA1Gl9YPwt5SVFwhys+SOjQu32a45LX7HPJmwXmNh9qnJuIiMjl6vUPoqSxC/+zo8wmj2fEe7B2dgqunpWMWE+E6pQlpCgsT3Mu5+w+Mcpty19zPf7qlh5b+lpERORysPyisqUbbxc34s0TDTYn5or8JNxQlGYDM65wJWUJLQrLQSDW68KinDjrjsEdXHlTV+ARERGRS8PVYHeWNOO5fdXodj6fmRyNa+ekWrs4kVCksBwE4iIjsDQ7EZFuhuUeqy3T4iQiInI5dpxushHlsqYuG4T5yKocrJ6ZiHCNKEuIUlgOAqxbLsqIRazzsad/AHUdPtS2+2wWs4iIyGj0Dwza1clXj9Rhb3mrDcQ8sCIHK/ITkRztgaKyhCqF5SDAWcmpsR5r6cNRAC5MUlzXaTs+ERGRi+EE8ZbufrxyqA67SpttVdgFWXG4b3kWchKj4HYpLkjo0rs/CHBWMndkczNinbN/ty1Feri6DX1+hWUREbm4rl4/jte24/Fd5dZViYtdXTc3DSvyEu3qpUgoU1gOEmFOYJ7nhOWUWE8gLLejTyPLIiIyCifqOvCrPZWoaO6BO3yGtYi7Z1kmZqhHnIjCcrAId9LyvMw4pMZ40N7jx6mGTvvINf1FRERGUtrUhXdONmGrc+Mgy62LMrC+KBVxke7AFiKhTWE5SPDsPzN+aHESlmQ0d/VZC7kun1bzExGR33bmzFCbuK0nG/FWcQPanePFoqw4rJ+biqL0WLh4yVJEFJaDBXdprCvLSvAiLdZj9cpHatrR0tM3tIGIiMhZOAn8WG07Nh1vwNGaDqTHeXD3kiwszk5AQpRGlUWGKSwHGc5azk+OtktprFtu6VJYFhGR9+OocntPP366vRz7K9sQER6GBZlxeOCKbBt0EZHfUFgOMrmJkShIjgqE5TYrxxARETlbTVuP9VPeXdaMDic0L8+Lx6euykeMNwJhmtQn8j4Ky0EmMyESeU5YnuH8V9fRi7r2XrT7+gOPiohIqGObuCPV7Xh2X5UdI+amx9hy1guz4xERplggci79VgSZ+MgIm+jHfsusW65o6Ua9szMUERFh+cXJ+k68fbIReytaEeUOxzVzU3HVrBTEeSOsb7+IvJ/CcpBhC7mkaI8tf80aNLYE4vKlIiIS2thItKvPj83HG/DakTqEO8l4SU48rpmdYqPLInJ+CstBKCnabe1/2EKurLEbFS09gUdERCRUDQyesZC8/XQTmrv7rPvFp68qsMEVLT4iMjKF5SCUGBWBhVnxcIeHobq1B1VOWNbS1yIioaunfwCnGzrxyqFaHK/rQGacF/cuy8b8zDjEeiMCW4nI+SgsByHOZp6ZEm31y919A6ht70Fduw+DLFYTEZGQMjh4BrVtPrx8qAb7KlvtvqW5CbhtUYZdiWT5noiMTGE5CHlcYUiJ9VgbuUh3OBo7+lBc3+HsMAMbiIhIyGBHpENVbfjFu+Vo7e630WR2v5jnfGS5nohcmH5LglSEswNcmB2HxGg3Gjt7cay2AwMaWRYRCTl7y1vxwoFqNHT02Uqv1xel4vp5qYFHReRiFJaDFOuVF2TFW/1yA8NyjROWNbQsIhJSTjV0YuupJuwua0VY2AzcuywLq2cmq05Z5BIoLAepiPAZmJcRa/VobEDPfstcza9/QIFZRCTYcY4KJ/W9ebwBO0qa0OsfwNKceFw/L83mtKhOWWT0FJaDlCs8DLlJUUiP9cIVFmZBmSMMnPAnIiLBzecE5RO1HdYq7nRDF3ISI/GhZVlWpxzrdQW2EpHRUFgOUhwzYClGfnIUshIi0eOE5H3lbWjr0dLXIiLBjiu3PvbWaZyo74A3IhxLchLw4BU5VponIpdGYTnI8XJbXlKkjSjvrWhRWBYRCXJctfWt4ga8W9KMTp8fVxYm4UPLsuFxhSNMi4+IXDKF5SA3MzXayjFYr3asph1Nnb2qWxYRCVIcGDlQ2YYNh+vQ2Nlnq/NdPSsFC7PiFJRFLpPCcpDLio9ETmIUIsLDUNfRayv6tWt0WUQk6LA76Mn6DrxzshF7y1sQ43XhxnnpWFmQaItUicjlUVgOctEeFzLjvchOjMQM579TDV2obvMFHhURkWDAoMyrhlzO+q3iRhtFXpAZZ90v5qTFBrYSkcuhsBwCUmM9ttNkq6CT9Z02uiwiIsGju8+P14/WYWdZi/XWz02KxB9eNwt5SVFQ9YXIB6OwHALODsunnLBc1dKDgUGt5iciEgzYJq6sqRvP7K2yAZHshEjcOD8di3Pi1SZOZAwoLIeA5Gg35qTHwhsRZktfV7f1qCuGiEgQYPlFTZsPW040YGdpsy1Gsjw3ATc5YZl1ylp8ROSDU1gOATGeCGtInxrjsQbMtc6OtbSpK/CoiIhMVz39fhyobMVTeyrR2TuAuensfpFsq/Wp+4XI2FBYDgHcX0a7XViWm4A4b4SNQnBlJxERmd72lLVi8/EGVDZ3w+MKwx2LM7F2Toqz31dQFhkrCsshItIdbmGZl+UYlo/XdeAMr9+JiMi0w3ILXiXcElh8hAuO3LM0C8vzEpAY5Q5sJSJjQWE5RDAsLwmE5dbuPpsM0u7za6KfiMg0w3GOnr4BvMk6ZScot/v6MSc9BrcvzkBBcrT11ReRsaPfqBDBy3Oz02KQEusB83F9R68TmLvQ69dqfiIi04nPP4CK5qHuF+ydz8WnritKfe/qoYiMLYXlEMGJHt6IcMxMiUaaE5g7fX6rdevq9Qe2EBGR6YDtP3+5pxIn6jusHGOpE5I/sirPFqESkbGnsBxCON1jTloMshIi0eHrx57yZnQqLIuITBts/8mlrDccqkV7jx9rZ6fg5gXpSIpyq/uFyDhRWA4xs1NjrGF9V58fh6rard+y6pZFRKY+v7Ov3l/RijeO1dtE7dzESAvLbBPnCldQFhkvCsshJi85GlkJXpsgUtHSY7OpVYohIjL1VbV04+2TjdhR0mwlF9cVpeGK/ESkxXkDW4jIeFBYDjEJURE2GYRLYHNEmS3kONlPRESmLtYmv3K4zoIyO2Fkx3vxwBXZmJseE9hCRMaLwnIIykjw2ipPZ5z/jtW0o67dF3hERESmGl79O1jZhk3H6lHa2IWClCj8zrqZNv/EFabDuMh4029ZCMqMj8S8jFg4WRnHajus9o2jFiIiMrXwCmBVaw8e31mB4vpOW3BkdUGS1SrHely2QquIjC+F5RCUHuvBnLRYRLjCLChXOztizqoWEZGphWVyu8ta8MrhWvT0D1ibuPVFaUOjylp8RGRC6DctBCVEuZGXFIXkaA8GB8+gsqUH5c1dgUdFRGQqYPeLAxWteOlADZq7+pAS48a6OSlYOzs5sIWITASF5RDEy3ZxURFYWZBoM6oZlHl5T0REpo5Tzn5566km7ClvAUuTH1iRYyUY7vDwwBYiMhEUlkNUnNeFVU5YjnE+soVccV0HzqhuWURk0nEOCTtevHakFjtKmqzcYt2sVFxZmGzlF6pTFplYCsshiiF5WV6CheaWrj6UN3ejpVsLlIiITDYG5T1lLXjnZJPNKclNisJ9y7MxKy0GUW6NKotMNIXlEBUZEW5LX6fGeGyBkvr2Xpxq6ETfwGBgCxERmWj+gTO2WNSTu4a6X3BgY3luAm5fnIHEqIjAViIykRSWQ1TYjBlW91aUGYu0OI+NKnMkw9c/ENhCREQmWn2HD9tPN+HN4gZnv9yHlflJeHBlDjyucNtvi8jEU1gOYdzvLsiMQ2a813bKeytanbCskWURkcnQ2z+A/c5++Om9Vej0+bE0Nx5Xz07GrNQY1SmLTCKF5RA2w9n7zsuIQ4YTlrv7/HbJr6mzF31+BWYRkYl2tLbDul9wsSiWX1w3Nw1X5CcixuMKbCEik0FhOYRxoCIvOQpZ8VwydQYaOnpR2tSFDl//0AYiIjLu2P2i3dnvbjnRYCUYtDQnAetmp6AwJcb+LCKTR2E5xHGiX05ilM229g8M4mBVGxo6+wKPiojIeOPVvL3W/aIRpxq6kBrrweevKcTstBiEh6n+QmSyKSyLBWUuf81Z2Acq29DY0Rt4RERExlP/wCDq2n34xbsVOFnfifzkKNy6MAPzMmNt0SgRmXwKy+KE5UgbweDSqtxZ1zo77l6/umKIiIy3mjYfXjtSh/2VrRg4cwZLcuJxy8J0JES5NaosMkUoLAvSY70oSIlGlCcczV19qGzpRpNKMURExlVnrx+Hq9vw/IFq2/fOTo3B1bOSLTBzHomITA0Ky4JIdzgy4rw2kSQiPAyljd0obeoOPCoiImONa6XySh67X7D8jdH45gXpuHZumhOUdWgWmUr0GykmISoCy/MS4I0IQ0lTF0oaOwOPiIjIWGJQ7ujpx6Zj9XjjaL1NtL5tUSZW5CciJcY9tJGITBkKy2ISo9xYYWE5HJXN3Ta6zBna3KmLiMjYYeehTcfrsbO02VrG5SVF484lmbb4CK/uicjUot9KMXFeF+ZlxDqhOcLq6KpaulHT1oPBQcVlEZGx4usfQHlTN145XIsTdZ1IjvbguqJUu7LHQQsRmXoUlsV4IsKRHue1NnKsYa7v6MWh6nZrayQiIh/cmTNAY2evjSrvKW+1lVMXZsXhIytzER8ZoSWtRaYohWV5j9sVhuW5CdYdg2F5X3mL9V4WEZEPrqffb0tZ/3hbGZo7+7DM2d/eMD8NOUmRcKn8QmTK0m+nvIc762W5iUiN86Cpqw8Hq9rR2ee3pVhFROSDYdeLVw7V2iIkSTFurJ2dgjUzk61OWYPKIlPXuIXlM07Aam5uxrZt2/Dkk0/ihz/8od2eeOIJbN++3R47V1VVFTZv3owf//jHeOyxx/CLX/wC77zzDlpbWzE4qHKA8ca+nlycJDPOa/9+1W09KGvqRnfvbxYoYW4e7sVc3txtO32WaihOi4iMrKa1BztON2N7SbMF47Wzk7F6ZhKyErxDG4jIlBX+t47A52Oqo6MD+/fvx09+8hO89tprePvtt7Fz504cPHgQlZWVSEtLQ0pKCtzuoQkNDMRvvvkmHn/8cbz44osWqPfu3Yv6+nrEx8cjIyMDERGs6Rr9+Xd/fz82bNhgH++55x77uy7l+aGGP5sod7hdJjzd2IWevkHkJEbaja2N2nr6LSTvr2jDoap2FNd3OGG5d2jk2fmxMmzrUqKIyG9w98iV+TYfr8erR2pRXNdpAfn3rp2FxTnx1oFIRCYOM+GWLVssiz788MOWDcMu0tt83JLNvn37LPi+8cYbuP/++/HNb34T3/jGN3Dvvffi1VdftdFmbjNs48aNNpJ8+vRp/Omf/im+9a1vWcA9deoU/vqv/9pelN/vD2wt44WBme2LeOvpH8Cu0mYbSa5t68GLB6rxpZ/vwV8+cxD/d8Mx/MurJ/A3zx3CHz2+Fz/eWmrhWUREfmNgcBCtzj70+f012FfeZhOp71+Rg/mZcYj1RAS2EpGpbNzCcnV1NVpaWrBy5UqsW7cOixcvxpIlS3D99dfbx/b2dtTW1tq2vOTPUeWBgQELyDfccAOWLl2Khx56CLfddpuNLrMco66uzraX8TUrLdrKMXqdsHygqg07Sprw9N4qm5RS3tyD9h6/BWm2QOruG0BNqw8v7K/G/2wvx4HKVuvPLCIiQENHL360tRQn6jps0ScuZX33kixrE6cLnSLTw7iFZY/HY2UTPT09NlrJP/NGPp8P0dHRiImJsYDc1taG4uJixMbG4sorr0R6erqVXsycORMLFiywco0jR46gsbHRni/jKys+EvnJQy3kmjp7sfl4AzYeq7elWRmQz53w1zcwiMqWHmw/3YSXD9Wiq9evGmYRCXm8Kre/ss3KL5o6+zAvIw43zktHnrN/ZfchEZkexu23taCgAEVFRTa6zFFjlmOw1IIfu7q67LHCwkKrHeGIMQMzA3J+fn7gKzjfXFjYe/dxpJp10CPhBEAG73NvHLWWSxMXGYHsBNYqR1m9HfuBHq+9cIkFf8o1bT5bvpVt5/o1uiwiIcw/eMYGGLhPPFnf5exXXbhqVjLWzUlBmIaURaaVcQvLy5Ytw3333Yfly5fj7//+761WmbXLX//6160s49Zbb8WcOXOsDpmBmqE5MjLSRpfPxsLruLg4K9vo7e0N3Pt+DMp8jGGaoXv4xuf09fUFtpJLkRbnxbLceGtpxNHk3lGEX27DoHy8rgPtPf2Be0VEQk+Hrx/vljTjhQPVOOP8d11RmoXl1NihK6wiMn2MW1g+efKkjSSzq8WnP/1pm+DHGz9//fXX8dxzz+Ho0aOBrS+Mo8MX6mLR0NCAn/3sZ/jUpz5lNc683X777RbQX3nllcBWcilSYzzWc5lh+VKwRINBmaUZIiKhiMcsLme9pbjBrroVpsTgxvlpmJseM7SBiEwr4xaWd+/eba3iOMLMSXt33XWX3T70oQ9h/vz5OHDggLWGCw8Pt9Fkl8tlo8OscT4bR555H0edWQN9PnyMZR0333yzfX3e+HfeeeedyMvLC2wllyIhKgJz0mIQG+mylnCjNcP5j7V4uswoIqGIV9iO1nTgnZNNVoaRHO3Gh1fmYH5GHGLU/UJkWhq3sMyWb+Xl5Vi9ejUWLlyInJwcuy1atAhXXHGFTdYrLS21kMwJfJzw19nZ+V6HDOLZOe+rqalBamqqbXM+DNtXXXUVPv/5z+MrX/mK3f7kT/4EX/7yl+3vk0vH3p/pcR7MTI5GlMcVuPfCwp1QzT7NmfGR9lFEJJQMDJ6xSX0bDtfiYGWb3bckJwF3Ls608ouJHkPgnJN2X7+Vx3EBqRbne/MPaB6PyKUat7BMnGDHzhccMeYIMW/8nDdO3uOoMkeLuUBJZmamLUxy6NCh97bnREAG5ZKSEsyaNQuJiYmBr/x+LNHg1/F6vYiKirIbR5t5YxiXyxPpdmFNYRJSoocWjrkQHgOinICdmxSJ2anRiPVqBEVEQkt3nx8n6zrwqz2VtgLqvMxY3LYoA1mJkRPa/YKhne09GdwPV7XjnZON2HKiAbvKWlDjfF+dvc7x2NlGREZn3H572cGCI76sJWY3jBMnTtht06ZNeOqpp6xtHAMwMeyuX7/eAjIfe+utt3Ds2DH88pe/xLPPPmtf55prrrFALROHreNWFyQhKebCYXl4sGSmE5I/sjLXummoCkNEQgmvhB6r6cD/vFtuITUt1oM1M5Nx7SR0v6ht9+GlA9X48i/24H8/fRD/9+WhRaT+9rnDgUWkSizUi8jojNty1xzV5UgvR4bZQ5mLijAEc1If28HdfffduPrqq5GQkGDb8yO7WlRVVVmtM7flKDM7YbCrBjtoMDRfaKLfubTc9QfDHXy0Jxw7S1tQ0dwDDkQwQPMjL+8Rf5oxHhdWz0zCXUuysL4oFfGRbtUsi0hIKW3qxqbj9dhwqNYWa2Lpxa0LM1CQEj2hx53y5m68fqQOT+2qxBEnvDc5wb3D57f+9/zY2s0VWX3ocr7HtDiPDW6whE4kVFzOctfjFpYZiFlnzBFk4s6CATo3N9dGkTlSzM+HdyIsseBzWELB0g1+85ycx0DN7hbJyclWtnEpFJY/GAZe1i5zsgoXHeHEFS5Wwol/+cnRyEmMxOz0WAvKNy9Ix9WzUpCdGKWgLCIhg+MGXO2UQfmVw3VOaO5CUUYsHroiF8vyEuFx9qEThYMYm47V4fkDNTbI0T8w+N7AxjCWX3DkmyuxsjRkXmYcIifwexSZbJcTlmec4bWjIMSXxZrnRx55BN3d3XjsscdsgqDC8qV7bn81ntpVYStRrcpPxL3Ls5EU7ba6uPjICFsaO8rtcn62gSeIiIQIhs/TDZ34943FeONoPRKcfePvrC3ArYsykJsYFdhq/LFtp69/EP/fs4dsdJt1yRfidYJyYWoM/v3h5ShIjoYrXDtwCQ3MhI8++qi1NmapL7PhxQZjJ27GgUxbRemxNpLMy3i7y1rsst3yvERcWZiMBVlxCsoiErJ6+vz4xbvlNpjAUeSFzj7xjsWZ1hVoIvX7z6C4rgPVrT3ocr6nixleROqE85w2LSIlckEKy3JRLLdg+UWs14UOJzBzFKWlu88u4XHREgVlEQlFTV291mlix+lmNHX2WveLD6/MtStvl9KffiwMnBkqr+Do8miuF3MT/+CgPafXPzB0p4icl8KyXFS0x2WXE2emRNulvsPV7ahqef/iMSIioaTPP4jTDV14fn+1TarjSDKvtq0sSJzQNnFnC3MC+qUOXnBynwY8RC5MYVlGJTsxEktzE2yFviNOWObBgcFZRCQUsaPErtIWbDreYPvCKwuTsG52ChKjJqcbEEey2a6Ogxuj6W5hE7hd4bZYiib4iVyYwrKMSlZCJJY5YZn74OL6TlQ4YZlN70VEQg0nkG891YgXD1ajb2CoS9C1c1NtHzlZIlxhmJMei+wEL6Ld4e/1vz8fPuZxtmdQ5pwUtvsUkZEpLMuoxHtdyE+KsvpljlqUOWH5RK2a2otIaOEFNU503na6CaWN3Yj1ROBja/KxMCv+giO6HH3mktP7K1qx9WQjDlS2oamzb1T1xaPBv5l///VFadbOc/i+cw3fNzMlCh9emWMdjVSGIXJhCssyKq7wMCRFe7A8LwExnnArwzhS0x54VEQk+LFNXGtPH149UmfdL7hI07VzU7DGCaccpT0fX/8ATtZ32CIhj205jR+8XYIfbyt1Pp7G9946jWf2VaGksQv+gcHAMy4fMy/L5W5blInrilJ/a8ERfhblfM8rnP04O3ZwNJx/FpELU1iWUWM3DC48wpGIqpZuW9qVk1zGamRERGQq6/D1W53yW8WNaOzotR7z9yzNQnZCpJU1nKunbwClThB+fn8NfrS1BP+zo9z61nPxkmf3VeNnTmj+qXN75VANKlu58NMHL21Lj/Pimjkp+OjqPNw0Pw2rCpKwKDve2nwySHOV1ftX5FhYZktQDoSIyIXpt0RGjctac7Z3QpQbDc6B4mR9p7UdYvshEZFgxkWYOFeDI8OVLd1IifHYIk3XFaXZpLrzqW7rwWtH6/Afm07aSqjnzvPgsth7y1vxX1tOY/OxetuvjgUG5lsWZuCfHlyKv7l7Af70lrn4yk1z8Fd3LsD/fWCplY1wQRIRGR2FZRk19lROj/daC7nEaLf1Wn63pNl2+CIiwYylEm+eaMDBqjZbRpqjt7cvzrR2befDGmVuy9Zy7IF8PryXt67eATy1u9IGIMayhpnfG/fXK/OT3ltEyib/nf9bFpERKCzLqHEH63YCM3e47I7R0t2P7aebbGU/EZFgxaWjOanvtSN16O7z4wqGz1nJVsYwUu7kKPHp+i6UN/dcNADz6txpJ4yzbzMXNxkr/N68EeFWQhfrjbD65JHCvYiMTGFZLtmirHhbpKTd14+9Fa1DpRgDYzQcIiIyhTDoHqpqw7ZTjTbyy/KLmxekY0l2/AUnxzEs13f4bILfxQw6fwfrm9kto7m7L3CviEwVCstyyYoyYpGbFGkzw3lpkvV7HHkREQkmg84+jlfONhyqxc7SFhulXZmXiGtmpyAnMSqw1fkxJLMH86Xw+QfR69xEZGpRWJZLxkt6+UnR1neZk172VbShtk3LX4tIcOlwgvLGY3VWgtHY2YuC5Gj8/nWzkBHvvWjdLydER0acf+LfSFhPHKXV9ESmHIVluWRcJjUvOQoLMuNsEsu+ilZUt/kCj4qITH8cGeaVsyd3VVpf+VmpMdaKje3iRrM8dGZCpC3ixMGFi1UJc5/Knsh5SVFIi/MG7hWRqUJhWS4Ld+qc6Ees46tq6bGJLyIi0x3rlBmQNx2rtxX3uLDHFWe1iRvNJLk4b4SVrC3PTXDC8NBku/Ph/ezRvKYgEbOcIM4RaRGZWhSW5bKkx3kwx9mxc9SELeRKm7pQq9FlEQkCnIPB/scvHqyxEeb5mbG4sjDJPo4WyzTmOWH5jiWZyEmKssmA4efUbjCExzj70Fmp0bh3ebaVeVysvENEJp7CslwWjyvcLhcuyU6wS5IcXWbTfRGR6Y6jyW+fbLTV99hf/p6l2U5YTnGC7KUlWdY23zQ/HX9681wszI635bHPxhrllfmJ+PINs3Ht3DTrtCEiU4/Csly2pGg31s1JscuGJxs6caSm3WqYRUSmI05YZrs3TurbWdpsdcQPXpGDxU7Q5TL/l4q1yAmRblw9OwV/cft8/J8PLcJXbynCH1w3C1+7tQj/cN9i/MnNRVg1M1mLhYhMYQrLctkSnIPHqoJEJERFoKWrz0Zhalp91lJORGQ64V6Lq5G+fqTeul+w7/Hc9FjcZWUUkXCFX16S5fM4YrwsNwE3zEvDnc7X+9CyLPvIGuhFThDnwIMWCxGZuhSW5bKxBo8zwznj2+Xs6Kvbemx51z71CRWRaYbhuKSx05anLmvqtlVKr52TYiGXk/U+KI4ac2CBy0/Py4iz+mTrlKGMLDLlKSzLZWP9niciHMvzEq1NEles4vLXvaNYsUpEZCqpau22xUcO17TbCf/yvAQ8cEUO3C4dJkVCnfYC8oFwdjdbKmU7Ybmxsw97ylrQ1tNvtX8iItMBy8j2lLXi2f3V6O7149q5Kbi+KM3KIy51Up+IBB+FZflAOIGFLeRyk6IQET4Dde29OF7bYYFZRGSq46Tk3eUt2HS8Ho0dvchOjMQ1c1KxJCfeOmGIiGhPIB8IB104+sI6vKz4SHT3+7GrrMVKMkREpjJeATtZ14nNxxvsqhjbYK6fm4oVeYlI10p6IhKgsCwfGC9TcnSZDft7/YN4t7QJde0+WwVLRGSqYUhm54taZz/11O4KvF3ciOauPmsVx77IDMrq6iMiwxSWZUzMSo3BvMw49A8MWhlGZUs3OntViiEiU0+Fs3/6pROSv/g/u/HrvVWoau2xAM2T/H9+5bhzX6W1whQRIYVlGRMsxchPjkZuYpSNyJyo60RpU3fgURGRqYHt4V44UIMnd1bYqqMcUeZJPseR+5yPXI306T2VeHZflfN5hxZaEhGFZRkbbK+UFe+1nqSe8DDnINSO4jotfy0iUwdz79aTTdh4tA6HqtutbOzsLMzPff0DFqI3HavHmycanCCtsCwS6hSWZcykxnqwZmYSvO5wnGrotBGaXr96LovI5Bu0OmU/NjsBmKViIxmOxqcauvDakXq0qxWmSMhTWJYxkxzjwYr8RMR4XGh1DjDlzd2oaB6qBRQRmUzdgRHjmjYfei6ycBL3WD7nRJ81zMec53T2+oceEJGQpLAsY8YdPsMJzG4szo5HQqTbJs3sKW9RWBaRSce6ZIZfrjA6mjJkbsPnNHX1agl/kRCnsCxjhi3kotwurJqZZCUZNa0+7C5rsVKM0RycRETGC2uRT9R1WMu4SxEeNsP6yYtI6FJYljHlcYXZ8teZ8V4rxThS3Y76jl6bZS4iMhIrfXACLWuEW7r70OHz28juWOD+hwsl8eR9tCUVXJ000h1uPZc9rvDAvSISihSWZUxxedj5GXHITYy04NzU1YfdpS1odQ5+IiIjOTN4BtWtPivd2nqyEQcrW539Rv+YtG5r6uzDgco27ChpsjB+MRxIjowIQ4YTlOdlxNo8DBEJXQrLMubCwmZgQVY85qbHotPnt/ZLDM0iIudiIH67uAF/9ewh/PmvD+D/9/IxfOP1Yjz6wlH86VP78e03TuJQVdtlh2bWG+8sbcav9lRaD/iClGikxXosEJ+vumL4Pi6y9KFlWfBGhKsMQyTEKSzLmONxZWFWHIoyYmzW+d6KVlS39tglVhGRYY2dvXi3pAk/2VZmvY/3lrdaWze2nTxa2+481owXDlTjqV0VOFbTbq3fLhVHlN852YhTzteM9UTgjsUZ+PDKHKybnWIjxiy3GOac5yPauW9NYZJtd2VhMlzhSsoioU5hWcZFfnIUZqfFIModjpq2Hjv4sWZQRIQ4Tlxc14lXDtfitSN1qGvv/a0aZZ5gc5sNh2ptu4aO0V+hYhce1j5vPlGPXaUtNgF5ZUEibl2YgYdW5uKhVTm4YX4aVuYnYlF2PBY5J/gr8hJx/bw0PHhFDm5ekIGcxKj3hWkRCU0KyzIuYr0RyE+KRlFGLMKdg82BijacbugKPCoioW5g4Ax2l7fgjWP17y0Ecj58rNkJvY/vrEBFS/eoW1HaVa3yVrxd3IiSxi5kJUTi89cUWhkGl+a/e0k2/uWhpfjruxfgT26eiz92bn915wL80wNL8MCKXOQlRQW+koiEOoVlGTcclblyJi9jhmF/ZStONnTYKloiIqVNXaho6ka77+KlFaw15rwHllKM5goVR6irWnrww7dLLCgXpkbjpvlp1gM+2j00WY8Dxq6wMMxJi8GqgiSscfZVPLn3RISpRllE3kdhWcZNepwHS/PiEet12eVQHrQ4MiQiwn1Cu290S0lzbp9/4AyancA8mtZvZU4If/lQDY7WtFsdMttZ3rIww+qR2Td5GEMxJ/BxH8UbW8Wp7EJEzqWwLOMmxjn4FCRHY3ZqDFzOAaq0sRuHq9sDj4pIKLuc5hajeQpD+L6KVrx6uM5GrRdkxWHtrGTMz4jViLGIXBaFZRk3HKGJj4ywWedJ0R6UNXe/t/y1ijFEQluCs2/gSO9oRnK5BUeE+RxOGh4Jy7y4ENLbJxtxrK7Dtr1tUSaudMIyy8FERC6H9h4yrtiaaX1RKpJj3Kht89mBrLWnDwNa0U8kpOWnRNskujiv67z9js/GK1NJ0W7MSouxpfTPh32YuWroq0fqsOVEA2Kdfc9dSzNxRV4ikp2TdRGRy6WwLOPK7QqzUoyZzoGRNYGcnLPtVDM6RrnkrIgEJ3d4GJbnJeCaOSn255ECM+9PiHLj/uXZFq5HGiH29Q3g2X3V2FM2dPWKiyLdszQLuUmR76tTFhG5VArLMq54iZW1y1ykJCchEi3d/dh0rB5tzkcRCV2svmD3iVsXZVhgTolxI+KcBUC4ZD5PtG+Yl4o7FmfaqPL5Ym+Hz4/DNe147Ugtypu7kZMYiVsWpjv7nXjEeCICW4mIXB6FZZkQS7ITMCs1Bp29/dh+uslGmM9dgEBEQktarBdrZ6fgs+tm2sei9DgbPWbY5cJGXCzkNidMP7wmzz5nWde52FausqXbFi454gRmjiJzcZG7lmRazbIm9YnIB6WwLBNifmas9Tql2nYfjtd1oLFz9KtxiUhwSoh045o5qfiH+xbj/z6wBP/7jvn4yk1z8f/dtRDf+PAy/NGNc7DYOdkeqZSivafful88uavC+dxvS1RzZb6MOK/awInImFBYlgnhdYfb5VReFuWM9Z2lzShr0op+IqGOeZYT+KI8Lsx0TqhXz0zCtXNTsSI/ARnxXuuDPFJQ5r7knZONePFgDXr7B+2EfL3z3KU5Cc7XVVAWkbGhsCwTgiM8M1NjnANgorWN21/RitONXej1DwxtICIhjdGWZRPsesHa5MQot00QHgn7NLPsgmGZXXaiPeG4fVGGTRrkc0VExorCskwY1iEuyxla0a+q1WdL19a1+wKPioiMDrtddPj68frReuwpb7U/LwzUN+cnRalOWUTGlMKyTBguUFKYGmMz4DnL/URdB/ZXtAUeFREZne6+ARyv7bAlrU86J90s8frklfnIT46GJ2LkRUtERC6HwrJMKF4evXVhBhKiIlBc14l9gVEhregnIqPhHxxESWMnvr3pJGpafShIicK6OSk2sS9KQVlExoHCskyouMDy16xJ5GXUU85B71RDJ/r8aiMnIhd3uqELm443WAcMWjMzGdcXpdmVqzAtPiIi40BhWSZUZES49VvmjbPfq1t92FHSjJ4+TfQTkQtr6+nH7rIWvHG0Hu3Oyfa8zFjn5DsZi7LiAluIiIw9hWWZUJx4ExEeZu2hOBGHi5O8ebzeRpk5u11E5HwGnR3EsUD3i4NVbYhyu3Dv8mzrsKM6ZREZTwrLMuEYmFcVJKEgJdpGh3jg4xK13X3+wBYiIr/BeQ1cfOTXe6qw43QzkmPcuGNRhu1HUmO9ga1ERMaHwrJMCi42MDstBlnOxw6f3xYpqW1TGzkR+W0t3X34lROU91W2om9gEPMyYvGRVbm2/+CCJiIi40lhWSYFa5fnpsdiQVYc+gfOYPupJlS0dNulVhGRYaxTPlzdjhcOVKO6tQcFydE2oW9pboKVYoiIjDeFZZk0HFnmAY9YilHa2I2uXk30E5EhLL9g94uNR+twqKodrrAwXDUr2RYf4dwHLT4iIhNBYVkmTVZ8JOZlxCElxo2+gTM4VtuO4rqOwKMiEupYfrGjpAnP7quGf2AQ1xel4to5KciIU52yiEwchWWZNK7wGciM99qCAlHucByqbrcR5jMqxRAJeRxVfv1IHd44Vo9+JyjPSY/FjfPTMD8zTv2URWRCKSzLpOLiJNfNTUWs14WK5m4bXWY7uUHnQCkioanXP4gDla14q7jRrjYlRbvxoWVZWJKTgIQod2ArEZGJobAskyoh0o1luYnIToi00SPWJx6qakO/wrJISPIPnEGjc8L84oEau9I0Y8YMJyTH477l2dZFR3XKIjLRFJZlUrEUIzHajTWFyc6BMBKVLT22lG1vvyb6iYQidr/YX9mK5/ZXW4ecRdlx1iYuLc5rk/pERCaa9jwy6dxOYF4/NwV5SZFo7OzF3ooWaxHlU2AWCSlWflHVih9vLbXQzAnA18xOtatP4apTFpFJorAsky48LAxz0mLtFhcZgdr2XuwoaUZTV19gCxEJBYer2rDpWD2O1LQj0h3unESnYu3sFMQ7+wVFZRGZLArLMulYg8iQzAVK2Hu5u9ePzcfrbXRZi5SIBD92wGno8OGtk41452STdcJYnpeIa52wPCc9JrCViMjkUFiWKWNxdjyW5iTYpdgdp5tR3tytUgyRIMfTYfZZ5+/8O05YLmvqQmqMB5++qsBOoFWnLCKTTXshmTJyE7lISawtONA/OIj95a04UdcZeFREghEn87Jt5JO7KnC0ph35yVG4d0U2FjpBOdaj5axFZPIpLMuU4YkIR4FzoFxTmGTL2u6taMXR6naVYogEKf5uV7T04PF3y3GstgNeV7iVX9y+KAMJURGa1CciU4LCskwpWQmRWDc7BTFeF8qaunG8rgMN7b1QXhYJPvXO7/au0mZsOFyLzl6/tYnjpL656bEqvxCRKUN7I5lSkmM8NrKUFR8J/+AgTjZ04gCXwLbKRhEJFn0Dg3b16NXDdahu9dnKfDfOT7dJfWFaeUREphCFZZlSeNmVbaKuK0pFepwXJQ1deLu4QctfiwSZU/Wd1vVm2+kmuF1h+PDKHKyemWRL34uITCUKyzLlRLnDLSxnxXvR0t2HIzUdKGnsUmcMkSDAtnDtvn48s68a75Y0IzIiHGtnJ9uIck5ipEaVRWTKUViWKYejTHPSYzHbucV4XNZvmS2lWrv7A1uIyHTF3+M3jtXjneJGNHX2YXZaNO5bno3ZqTHOibJGlUVk6lFYlimHI0sMyctyE2yREi57+8rhOtS2+WxUSkSmp+6+AZys78BPt5WhrLkLqbFuXDUrGbcsyECsNyKwlYjI1KKwLFPWqoJELMqOR1efHzvLmnGqsdMu34rI9HSqoROvHqnF/opW9PYPWunF/Sty7GqSqi9EZKpSWJYpKzXWi6KMWBSlx9oEv22nmlCsRUpEpqW6dp+VXrD7hd/5fb5pQTqumZOKzDhvYAsRkalJYVmmLA9rl9NicWVhMmbMmIHdZS22wldPnyb6iUwX7JHe5x/EW05Q3lLcgKauPqtPvnlBGhZlxdliRCIiU5nCskxpeUmRtqIfV/OqaumxVb4qWroDj4rIVNfrH7DFhVh+ccQ52WUv9VsXZWBlfhLSNKosItOAwrJMaVyogGUYHIGKdIfjmHOw3XG6OfCoyOTjyCmXbeYtGFea/CCv74zzhKbOXvx4Wyn2V7RhcBCYnxGLz68rRGZCZGArEZGpTWFZprxEJzB/aFk20mI9ONXQhZ2lzTbRT50xZDIxOHb1+nG4ug3bTzdhV2kLSpu6gmYBHb4+9jY/Xttu/ZD5e3eyvhP+gdGvp1nV2oONR+ux9WQTWrv7sGpmIj6+Jh8xkS6Ea0afiEwTM5wz/3HZs2/cuBFvvvkmOjvPPyGrqKgIV155JZYsWYKBgQFs27YNBw4cQHV1tf05NzcXa9aswYIFCxAVFRV41ujxZXV1deGRRx5Bd3c3HnvsMURHR1vtq0wvrHds6PDhb58/gh3OQTs/KQp/dOMcK8+IU7spmQQsCTpY1Yp9FW2oaO62UBkWWH0yz3l/Xum8N1lvnxjtDjxjemGbxiPOScDu8lZ7fexIM8P5j6vr5Tqvb3VBIuZlxiElxhN4xm9jy8c3TzTgJ4FRZV4dYueLu5Zm2gmwiMhkYCZ89NFHsX37djz77LOWDcPDLzx3IvxvHYHPx9Tu3bstADP8NjU12a22thYnTpzA888/bwF47ty5yMjIwKFDh/DUU09h165dqKqqQn19vW3HsMsXkZOTc1kht7+/Hxs2bLCP99xzD9xut8LyNMQlsHmQLncO2lWtPrusO+CcDC3JTkBcpEv/pjJheDWDC2lsOl6P5/ZV45XDtThU3W4rTJ5u6EJxfQdO1HaixwnP7Buc5IRl7zSawMZBhuauPrx9shHP7a/GSwed11fVbq/ttPMai+s7baSZ/ZK58h7DMsujzsWf096KFrx8sAZvFzc6v6cRuHd5Nq4vSkV24qUPfoiIjBVmwi1btqCyshIPP/ywZcOwsAsXWoxbGcadd96Jb37zm/jhD3/43u3f/u3f8Pu///uIi4vDwoULMWvWLAvG//Vf/4V33nnHwvNf//Vf4+///u8xc+ZMvP7663jiiSfQ29trO3EJXQzEV81KwbyMWLT29GPz8QbUtPbYqLPIRGEI3lHShMffLbf3YIfPH3hkSP/AGTQ6J3NP7ap0wmaV1dhPJ2zptqe8xb5/LgTE0WHWKg9jCOYKfM86JwpP763Cwaq2wCO/we2bu/uw4VAtNp9osJMFjkQzKHNlThGR6WbcwrLX60ViYiKSk5Pfu3Hoe//+/fY5yyvS09NttPmll17CunXrcN9992HZsmWYP38+vvSlL1mYPnbsmD2HgVlC2/zMOLslREag3TmIbz3dZKPNIhOBmZFB8fF3K1DS2D1i3e7w/dtPN9sI9HSpYebr48nA03uckF/bcdHXt6+i1QIxA/TZ27KOmycTrHP29Q9iZko0Pn9tIQpTYwJbiIhML+MWljmkzRqQ4RtHBmtqaqw8Y9WqVVZawfBcVlaG5uZmC8+zZ89GZGSkBW2OLLNumSPKR44cQV9fX+Ar/zZu4/f74fP50NPT896Nf+b9Ehx42Xd+ZiyW5yXYCN5bxQ22Ipgm+slE6Ojtx2nn/cZyBNbwXgjfkSzXYPlCmXNC1z8w9a+AdDuvabjcouMiK2Xy9bV0O6/P2ZYlKL1OyCaWcLBbzRvH6lHZ2oPZqdF46IoczHKC8vnKNUREpoMJ64bR2NiI4uJiC8zr169HZmamTf6rq6uzx1m7zJHoYQzMHIFmLQnrSi4Uevl19uzZYyUbP/jBD967/fjHP7baZwkOLE2enRaD1TOT4I0Is5n5XKSkps0X2EJk/HT1DqDaCYAcOR3NCVqfE5BZMsTnTIcTOk5SrGzpsdISlmNcDE9YWaZR2dJtr5WLBR2v7cCz+6ttpc2kKLeVTl03Lw0xXhfCNLdARKapCQvLx48ft5IKjhxfddVVSE1NtdFiTuLjyDPvj4h4f2cDBmbe19HRgUE26BwBw/KOHTvw3//93/jWt75lt29/+9v47ne/a6PSEjwy4yOxODveOg7wYH2gss25tdolZJHxNODsgxgoL+W9xtDJ0obpcPGDgb6nz29X6kaL9cmc7Dfg/C5ysaB3TjbapEfef/WsZFw/LxXZCZEKyiIyrU1YWGZbOHa6WLlypdUqc8T4YoZ32hfrdsDg/clPfhI/+9nP8Morr9iNXTCeeeYZ3HrrrYGtJBiwM0ZWQpT1XeaqfmxJte1Uk41sXUqIEblUblc44iLdTvAL3DEK7vAwq7G/lOdMlgh+r1EXnxV+NpfzwtgGrt85kXjtSB2e2Vdl9y/MisfNC9KxPPc3VwtFRKarcQ/LHBFmTfLRo0dtkt71119vo8gMwB6PB7GxsVZiwfrlc+uSWXfMFh/snnGhHbjL5bJtWMrBWujhW3Z2trWek+CSHOPGDfPSkJPI0eVBK8fgohAc9RMZL/HeCCsD4vuPIfhCmI2j3OHIjPfaxDYG0aku2uPC3PRYpMa64XFd/PVxDkF6HF9fNF49XGdzCFiWkRUfiY+vycOi7HjnBGPqv24RkYsZ9z0ZFxg5ePCg1R2zJpkLkTAkE4MyAy6DMEedGaqHMSizzpkBOz8/3wLxhfBrcJtzb5cySiLTg9cVbmUYq/ITnQO7xzpivH6kziYlnd3mSmQsuSPCLPxeMyfVVpO82GAxgycno3LklVdEpjoGWwZllk+wdOJi33FBSpRNuOWiJez6caKu0/ou37E40+YVXGjBEhGR6WTckyRHi9n8mSPH7HbBG0MscTQ4Ly/PJvtxYZLDhw+jvb3dapT5eUVFhZVrsFPGcMAWYVUOD+zr2bc1LcZm5W891WRdBzjJSGQ8MDxycZx7lmbiCudEjYE5Ivz9kZKZ2Ou8N3MTI7F+biqucoLndCrX9TgnorcuzMAaJ+xmxHltBP3sb99+95z7spwwvbogCYUp0bZ4CcuhuB1/Lg+uzLGT2OlwgiAiMhrjGpZZc8yRYS57zWC8dOnS9y0pyMl7DMp33HGHLUrCZQf37dtnEwE5Qe/UqVPWc5nPU1iWs/FAzAPzvMxYW/SgvsOHLSca1BlDxhXfa8vzk/CZtQW4bXGGLW99diRkuUWaEzL5+H0rsm3J6+mEv1dLchLw8Svzce+KLJsXwGW8h7nCwmxVwodX5+LG+Wno7PXjFzvK0dTVi1VOwL5rSaYF6OlQdiIiMloznEA7btetW1tbra/yF7/4Rdx777347Gc/i6KiosCjQ9gLuaSkBD/96U9t8ZG2tjarc2aJBuubOUGPC5Vc6pLGfFnstPHII4/YqPZjjz1m9cuX+nVk6uI7d/Pxejy5qwKvH6232sk/vWUuri9K08FaxhVDYmNHr7VaY818u68frvAZSI3xYnZatHVtSXSCpscJ19MRey6zTzRfH3uZt3b32b4z2QnKrNvmSQKXxP71niocqW7Hwux4fOLKPNyyIN0mCYqITFXMhI8++ii2b99ug7TMhmcP5J5P+N86Ap+PueF2b2lpabj22mttOetzu2DwG2QtM29ZWVlWn8zR5LVr1773nHNbyo0WJweyKwY/3nPPPfZ3KywHD/5Tst8yg8ve8lYrx2CtZaZz4+iXyHhhGVC8Ewoz4r3ISvBipnOiVpQei6KMWBSmxFi5hmsan7DxZDPOCcR8fQz+Bc5rYg02XyNX5Ntd1mrdLw5VtyHW2e6jq/OslpvbiohMZcyELA/mXLqHH37YsuHF5reN68jyZNLIcmjghD6uGPZvr5/A/so2rMhLwIdX5uLupVmqmRQZY1ypjxNq/2PTKWw91WgnrOtmp+D3r5ttIZqt5EREprLLGVnWtWqZ1rjYATtjcAZ+jMdlgXlPeYutshacp4Eik4Mnpg2dvXhqdyV2ljbbSn/zMuLwxetnO7+DkQrKIhK0FJZl2uPMe7a7ynUO2Dxcs4XVWycbbaEEERkbtW0+vFXciOf2VaO+oxcrC5LsKg77nbsvMiojIjKdKSzLtBfhGupAwIl9rLE83diJN47Wob2735bwFZEPpr2n30aTn9lbZaPLRekxzu9bKlYVJFr9tqrbRCSYKSzLtMfjNFdLY1ienRqD7t4BK8c4UNVmXQpE5PLwVNPvnHDuq2zFG8fqcdD5vWKXD66gyR7SPEkVEQl2CssSFDh7f2FWHBZnx9tiEQ0dvXjhQDWqW3u0qp/IZeKVmfp2H17YX423ixutFd7K/ERbuISdP0REQoHCsgQNLp6wdk4K1hQmWzu5jUfqUVzXiU6fP7CFiFyK9p4+/GhrKd4taYHPP2grZn75xrnWKo+Ta0VEQoHCsgSVWc5BfFlugnXI6Orz28IJR2raA4+KyGjxqswrh+uw+XiDXalZlBWHj6zORUFylK1kKCISKhSWJajEeiOwIDMOa2enWJ/lXaXN2F/RahOURGR0Wrr6rAXj8/urUdbUZa3hritKs57KUW6XRpVFJKQoLEvQyU+Owo3z0pAc40Fla49N9iuu7ww8KiIX0jcwiMPV7Tahb0dJMyLd4bhhfroTllORHudV5wsRCTkKyxJ0uPzu3IxYrJ2djORot3Pgb8OrR2oDj4rISDgZtq7NZyPKLMGIjAjHdXPTcPOCdE3oE5GQpbAsQYcDX2xvdf/yHOQmRlm9JUsxDlW1obtPk/1EzoedL1h+8WNO6CttRkT4DCzJicfDa3JtLoBKL0QkVCksS1DyuMKtldySnAQnOLtR1tRtreSaOvsCW4jIMHZXrGv34aVDtdhS3Ihm5/ekKD0W96/IwcLMeMR4IgJbioiEHoVlCUqc3BcXGWELJ3DCX1tPPzYcqsPphk509Wp0WeRsLd1DE/qe2lWBypZuZCZ4sX5uKm5dmI4oT7jqlEUkpCksS1C7qjAZy/MS7BIyZ/VvPdVkH0VkCOuU91a04Ln91bZCH0eZb1uUgXuWZVt3GZVfiEioU1iWoMZlsJfkJuD6eam2dO+rR+qwr6INvv6BoQ1EQtze8hZsOFiLHaeb4XaF4SOrcm1UmSthioiIwrIEOa7qNzc9xjn4pyElxoP6jl7sKmu2dnIioazXP4jy5m48t6/aWsSxdOlall4sysCstBgLziIiorAsIYAheWlOvNUve5wAwKD8zslGWxKbl6BFQo1/YBC1bT14+VANNp9oQGtPn9X2378821bqi/NqQp+IyDCFZQl6rLnMiPfi42vykJUQifKmLmw/3WST/fr8g4GtREIDTw9buvuxp6wV39182gnNPhSmxNhCPrcuzLA6ZRER+Q2FZQkJ0R4XluYkYFVBIjLivNZK7mfby6wLgEgo4Qni28WN+NHWUrT7+q02mRP67lqaaWVLIiLyfgrLEhI4usxle7kS2YKseAsJO0tbbLGS5i4FZgkNLDt641gdNhyuRUljl02AZS9lTuhLitaEPhGR81FYlpCyKDseK/ISbHS5urUHG4/WW2gYHFTtsgS3nr4BOzl85VAt9lW0wBMRhpuck8ebFqShMDXaJviJiMhvU1iWkMKJS1fkJ1r/ZebjTccbrLcsR5pFglWvfwAVLd14YmcFtpc0O8F50CbyfX5dIWalxtqKlyIicn4KyxJyFmbFW4uspKgItPf0Y9vpRuwuawk8KhJ8ypu68eKBGrx8qBYNHb1YWZCIT11VgLkZsYiMUFAWEbkQhWUJOV4nHMxOi8F9K3IQ43Vhb3kr3ipuRH27DwMqx5AgU9rUhY3H6vHs/iprl3hVYRJuX5yBZbkJcIXN0FLWIiIXobAsIYfhgK3kbl6QZguWcDU/LvfLkgyt7CfBgpP5mjr7sNl5X79+pA41bT47SbxlYQbWzExGYrQ7sKWIiFyIwrKEpBiPC/Mz43DNnFTrvcxWci8cqEZVS48Cs0x7DMqdPj/eLW3GhkM1OFzTjsw4L+5YlIFrnfd8XlJUYEsREbkYhWUJWSzHePCKHCzOjrdOAYer2+1ydV17b2ALkenJ1z9o5Rf/sakY+ypakRAZYStYfubqAuQnKyiLiFwKhWUJWey9nBTttsl+VxYmo8Pnx6/2VOKIE5q7nfAsMh3xysiesmZ847Vim9jHDjDXFaXh01fPtMV5ZqhIWUTkkigsS0iLCA/D8rwEXD0r2YJzZXMP3ipucAJzW2ALkenDPzCIXaUteG5/DfaUt1iBPhcc4YS+WanRcDnvdxERuTTac0rIy06IwsqCJKyemYgw5zdi2+kmbHduXApbvTFkumAnl2O1HbY63xbnhI9/viIvEXcszsTK/CQ7MRQRkUunvaeEPF6VnpMegw+vzEVSlBvlzd14N7AUtlb2k+nA77xP23r68Yt3y7HpWL19ztrkL6wvxCrnRJDLWouIyOVRWBZxRLtdTmCOxZ1LMpGdEInD1W349Z4qCx3qvSxTXWVLN7635bT1C2/u6rPV+R65aQ7mZcQhyqOgLCLyQSgsizjCw2YgMcptYXl+Ziz8A2dwsKoNz+6rQn2HumPI1HWqvhOvHKrFK4dr7b26NDce9y7LttKLWK/LJrKKiMjlU1gWCXC7wqz38tpZKShMibbg8dz+auuOwWWxZWo7cwZo7e63xTeqW3vQ2NmL/oHBwKPBiX3BWZ/MoMxe4Xzf3rwgA+uLUm3CKk8CRUTkg1FYFglgrOAkqOvnp+Gauak2IrevvNXCyKmGLlvoQaYeXgXo8PWjygnI7ADx5vF6vHGsHttPNaGksQstXX3o8wdXaOZ7savXjzdP1OOlgzU4UNlm4fje5dm4cV4achLVS1lEZKwoLIucIys+ElcVJuMGJ3SEhc2wMPLOyUa09/gDW8hUUtPWgxcP1OBLP9+Dv3z6IP5xwzH8y6vH8TfPHcYf/WIvfry1FCfrOwJbBwe+FzcerccTuyqx3wnKmfFefHxNHm5dmIEcrc4nIjKmFJZFzsFL10UZsbh9UQYKkqNtdb+3nbD86pFanNHo8pRS2thlJzM/3VaG43UdtvoiSzF4a+rqs5FlltI8sbPCJm0GwwhzXbvPrnb8aGuJ1Stz6erbF2fiQ8uykeGEZpdKL0RExpTCssh5cLLfktx43LU0E8kxbhTXd2Dz8XrrY8sV0mTy8bzl3ZJmcInywzXtdlJzbqlMrxOOGZjZJeLVw3Xo7pveVweaOnvtNT+3r8pGlBMDK1DyxK4wNRoel3bpIiJjTXtWkfNgA4GUGI/1Xl6UHW/t41gXylHMho5etZObZMM1u1xA5nB1e+De8+O/FOuZeWWgobNvWk764+tl0N9b3oqXnffglhONiPG4cO2cFNy+MAPLchMCW4qIyFhTWBYZgTs8zOqX71qSiaU5CdZl4Wfby3C0ph2dTlCTycNyihN1Hdb1gqH5YjjCzO4mJ2o7rHf2dMPXe7SmAz913n+vH6u3RUaucYLyw6vzsCI/0Tm5U+mFiMh4UVgWuQBmkNUFSVg7O9lWROtwgtnj71ZgT1mLrZomk4Mj+1yOnCF4tLgaY6vznOlWt8xwv9t5v/37xmLr/c0SIXZr+b1rC1GQEq32cCIi40xhWeQikmM8WFOYjOuL0qy13P7KVrx5ogGHneAik4Mjqfy3uLQFN2bAdcnPmVwM97tKm/H4zgoLzBHhM3DVrCTctzzbeoJHe1yBLUVEZLwoLIuMQlF6LG6an47F2fHw+QfxzqlG6+XLCVeqX5547PiQHudFjNc1qu4PDMiREeH2nEj31F/+mfMUWeqzr6LV6uRfO1JnI8hcle+ORZm4dm6KnSxoTFlEZPwpLIuMgtcJWhzJ+8PrZllPW7bs2nS83kJzd5+6Y0w0rrY4Oy0GuYmRtqTzhUIjH4uMCENGvAdz02MQ540YemAK6x8ctA4sHFF+4UANwp2wvzgnHp+4Mt/6f7vCtOsWEZko2uOKjBJDGbsO3LM0C3PSY3G6oQs/3lqG47Udo5pkJmOLC8YwOK4qSLI/ny8wD983MyUaH1mVa0F5qldhcET5UGUrvrXxJHaWNNvo+ZWFSXjkxjlYmBWHcAVlEZEJpb2uyCjxMnhcZARuXpCOK2cmWYhhR4andlfYghia8DexmHmX5CTgziWZuGVhOuKdf5uzSzL4GWt61zj/Vh9aloW1s1PgneIlGFxMZVdpC/57a6kttc733JUzk/GxNXlYkBnnnLBN/bAvIhJsFJZFLgFrX7m63/qiNFyRn2gBmbXLXOGvvKkrsJVMlNRYD66elWKjxlxAZt2cFKx0/l1W5CXiysJk3LYoAw+uzMHNXAY6MWpKr27Xal0vmvHCgWq8fqTe+kOvLEi0kwEuPBLlBH8FZRGRiaewLHKJGJh5WfyuxZnWh5mjgVwdjh0yevr8FnJk4jAwXz8vHX93zyL8xe3z8Uc3zsGXbpiN/3XbPPztPQvxwIpcW7Z8quKCI2yBx9KLX+6uxPP7q63EhJNJH7wiFzctSNdkPhGRSaSwLHIZ2FlhaW4Cfu/amUiMjsDJ+k5w2eVNxxvQP836+AYDBkmWLLA2mTXMHFVm2QIX75jyNco+P3aVNOM/N5+2KxQcQV49MxFfvnG2vZYI1SiLiEwq7YVFLgP7/HI5bIayOxdnIi3WgyPV7fj1nkrrw9zhBCCZeOySwdZwDMmeiKnfU7mu3YctxQ34zzdP4VB1my1hvX5OCn53XSHmZcTZn1V6ISIyuRSWRS4Tg1kml8N2wjJHAFkPu6usBU/vrbK2Xz1qKScXUNvmw9ZTTXh2XzW2OR8Z8DkJ8e6lWVhTODSBVEFZRGTyKSyLfAAMzCvyE3HzwnQsz0u02tNf76nClhMNqGjp1oIl8lv4nuDKfFtPNzknVpXYfLzegvFVs5Jx//JsWymSfZSVk0VEpgaFZZEPiCUZ6+em4u6lmZiTFuOEoUE8ubMCGw7VoqO3P7CVyJC2nj48v78GP3y7BNtONVs7QvaL/ty6QqwsSLLJfSIiMnUoLIuMAa8rHFfkJ+HTVxUgNzHKapbZUu7JXZXo8w/iDNcvlpDGtwD7cv9iRzl+/m45Tjd0IjcpEvcuy8bnrynEzJQou1IhIiJTi/bMImOAtaWc5MdL6fctz0ZGghelTV14/UgdXj9ah+bufrWUC2EsvThY1YqXDtbgxYO1Fprzk6Nx28IM3LU0C/MzYxHldgW2FhGRqURhWWSMsHVZWqwXD63KtYUy2MngWG07frqtFAcqWtHmBGYJLRxN9vUP4GRDJ144UGPlF6c4opwYhdsXZeCepdlYnptgpTwiIjI1KSyLjCFX+AxkxHnxsdV5uK4oFd29A9h2uhk/216GfRUtKscIMf7BQdS19eLbG0/i6T1VKGnsRFqcB39w/SxbdXBeZmxgSxERmaoUlkXGQV5SFO5cnIUPr8wF52vtLm+xsMQ6ZgXm0NDe029dUf7+xSPYXtKErj4/Vs9MwiM3zsV1c1ORFO0ObCkiIlOZwrLIOODCGPMz43D74gzcvijTFsd4t7QZLx6swbslzehWD+agxXMh9lB+5XAdnthZgZ3Ov3tv/wDWzU7Bgytycc2cFFuim0tYi4jI1Ke9tcg4SYiKwNKcBDy8Os96MPv8g9h6kr11q3CitgOdvVrlL9iwz3ZlS7ddQeBqjlx0JMIVZis93r8iBzfMT0N6nHfKrywoIiK/obAsMo7YQ3fdnBR8ZGWuTeRq6uqzsPzcgSprHebXoiVBwz9wBjWtPdhwuAbf2XwSO8uaEe1x2eqOf3rrXFw/T6UXIiLTkcKyyAQYbinHS/B9A4N4Zm81fu2E5oOVbYEtZDpja7jdTjj+wdsleGxLCerae1GQHI37V2Tja7cWIT8pGu7w8MDWIiIynSgsi0yAWO/QCCNX+VszM8naiW061oBf7am0GmZevpfphyG5w9eP5/dX20Ijm4832NWDZbnxeOiKHNyzNMsme3ojwq0Xt4iITD8KyyITJCPei6sKU/DAihwszU2wZY83H6/HM3ursJ99mHvUh3k66XFOeMqauvHSwVr8cncl3i5utI4XVxUm26p8ty7MwLyMOOu/LSIi05fCssgEYmDmaOMDK7KxMCsOzV19eG5/NZ7cWYGjNe3o1qS/aYFBubSxy1Zn/NdXj2NXWYtN2luZn4jfXz8LH1qejYKUaI0mi4gEAYVlkQnmdoXjjsVZ+NjqfFvpj6ORLx2qsYVLtp9uVh/mKY7/PrtLW/DDt0vwn5tPWtlFUnQEHrgiG39yy1ysmpmEaC1dLSISNBSWRSYYRxuj3OG2QMWDV+TgzsWZYDzeUdKMp3ZX4PkDNVbTrMw8tfQPDKK6tQc/2lqK/36nBG8VNzr3ncGSnHj83jWFuJejycnR8LjCNKIsIhJEFJZFJgkXplhVkGhdMriiG/MVL+ezhvnlg7UWzPxOQJPJ19rdh73lrXhqV4W1/ttZOlR2we4mH12Vi5sXZmBOWqxN5BMRkeCisCwyiZJjPFg7OwWfuDLfRpppx+kmPPbWaVsqudIJzH3qlDFp2O2ioaMXu52TGHYu+eHWUpyo67QFZ66dm2ILzvDqQHZCpCbyiYgEKYVlkUnG0Uiu8Pb5awpx28IMK8k4VtuBf3+j2Los1Hf4hjaUCTV45gw6e/ttJPkbrxfjV86/RU/vADLivPjddTPxpevn2IlOeJh2oyIiwUx7eZEpgKOSs9Ni8OGVOfjSDbORleBFh8+PF/ZX41tvnLQSgO4+dcqYKI2dvdh0rB5/99wRPLmrwlZbTIv14q6lmfibuxfgpgXpSIvzaDRZRCQEKCyLTBFcGnlOeqz15/301QVYkBWHth4/3jzRgB9vK8UbTniraumxEU8Ze/yxchLf3vIW/HpPpS0ywp99TasP8zPjbDW+h67IxZrCZGTGR8Lt0u5TRCQUaG8vMoWwJGNmSjQ+tjoPdy/Nsk4LrFnmCnFP7arEpuP1ONXQhV51yxhT7D5S09aDHSVNVnbx+M4KvH2y0QLxivwE3L/cCcorc2zZcnYyUbcLEZHQobAsMsWwywJHmTl57DNrC2x5bN63pbgB39l8Cj/eWoqq1h70+hWYPyj+/PyDZ1Dd6sMrh2vx5786iCecoMyV+TJivbh1QQb+5u6FTlDORX5ydOBZIiISShSWRaYo1sMOrwj3RzfMth6+bGH2qhPq/urpQ3jxYI2NhsrlYUhu7PThiXfL8Y8vH8Njb5WgvqPXTlTuWpKJ/3XbPPze+kLkJkYiIly7ShGRUKUjgMgUxSv9sd4IzM2IxR2LM/G5dTOxdlaKPbavshWPv1th5QLbTjfZMtkaZR4dhuTy5m687Jxs/PvGk9ZxZFdpMzp9/ViWm4DPXF2Aj67Kw5WFSchKiIQnQmUXIiKhTGFZZIqLdMJaYVqM1TA/cEUOritKQ2a8F/udwPzcvmr82gl7nPx3qqETXQrNI+ICLzVtPuwsacaLB6qtywWD8unGLqTGeXDD/DTrmcwbl6xmD2wRERGFZZFpgAOb8ZERuGVhOn5/faGt+peXFIWGzl48t7/aygjYweF4bQfaff1OMFRiHsYOFx3Oz4SdRDYfr8e33ziJ/9x8GttPN9sEvgWZcXh4VR6+eksRPrwy10aTXWoJJyIiAQrLItMIJ/rlOiH5o6vzbOLZ7QszkBztRm27Dz/fUY5HXzyCn24rQ1lTV+AZoY0r8J1wTiB+tr0cX/3lfvzb68XYU95i7feKMmLwyI2z8Vd3zsd9K7KRERcZeJaIiMhvKCyLTDOcbJbkBOSlOfH4xFX5+ML6Wbh5Qbqt/HeyvhPP7qvCNzcWW9eMIzVt6OwNrcVMGIRbuvusDdx3Np/EtzedtHZwR2s67GexKDsen3J+bn9801zn55aBWWkxNmrvCtdosoiI/DaFZZFpiCPMcU7AW5qTgNsXZVprswdW5GBeRhyanaD4+tE6PLW7wiYBvnSwBvsqWtHa028jrcGIAZm9klm3zYVEWJLCFnC/2l2Ft4obLTxzYZF7l2XZKon3Ls+22m+O0rO3tYiIyEhmnHEEPg8qfFldXV145JFH0N3djcceewzR0dGYoWntchG9/kGbDMZcyfZtHlfYuC1rzM4MXHSEIZZ/BUeNL2dlOIZFjpq+drgOG4/VY78Tjlm73N03YK3PuOrcTfPTMSs1BonREYhyu+x1jTe+Lv48P+jrGwl/dt1OSG7v7rdSlO2nm7D1VKONInOyI7uJpMS4sTArDtfPS8e62SlIiIrQMtUiIiGKmfDRRx/F9u3b8eyzz1o2DA+/8KCJwrLIWRjqShq7UN3aA59/AAmRbsxNj7XL9GP91mHAberss9HQVifsxXhcyEmKRF5iFMIuI8zxN7nPCfn1HT4cYKeMvTXYWdaMtp5+m7AW6Q7HFflJuHZuii10Mjct9rL+ntHi98O+0MX1fH19NoKbnRBpKxSOxd/L3/FS599qZ2kL3ixuwI7Tzc7Jgd9+Bhx5T4xy24TIm52TBJZe8OcbET5D+wARkRA25cJyX18fGhsb8dJLL+HAgQOoq6uz+1NSUnDXXXdh1apV9jnx2zh06BA2b96MnTt3oqenB8nJyVi5ciVuv/12pKenw+Vy2bajobAsl6Kpqw+Hqlrx2pF6mxzX3uPHgPMe4kgo27StdsLl2lnJmOME5w+Kgfx4bTvePNFo5RH17T70Dw4iPCzMCXgRmJMWYzXI8zLjEOeNCDxr9Nj9od0JyGyTxqC6s7TZ+giznjneCf/JMW5kxHmRnxxlZRxLchNs9JmLcYyVE3UdeOdko/3dNW299j2FO797cZEuC8t8fQuz4q32erT4O80Sk9LGbhypacfR6nZUtvSgzjk5aOzsRUtXv72uhdlxWO68Jn79zAQv0p37Yr0uC9AiIhLaplRY9vv9KC4uxtNPP42SkhK7z+12W+DlN3XDDTfgiiuuQGZmpj129OhRPPfcc9i2bRsSEhJs2/b2dtt+yZIl+PSnP43U1FSEOYFiNBSWZbQYtN4tabaexbvKmq229+zWa1HucAuw64tScd/ynKGFKi6zlIBlEQzjv9pdaWGypKnbam2HuZ1wnhrrsZX72KFhqRP6OEJ6uTiqzJB8qKoNh6vbbNS8vLnHSjSiIsJtchtLMwpSopCTEIUM58QgPc6DRCfERrsvPTz3Oq+luq3HaoW5PDfDeo/zmodxZJev5wrn9X1oWRZWFiQhZYR+xvw34PfPfx+Oltc54b/CCcdlzd32M+SS1BxJjvFEIMcJ+4Wp0ZidFoui9BjnY4xz39AIvX7jRURk2OWE5fC/dQQ+H1P19fU2Svz973/fwi5HhzmafO2112L27NnIyclBfHw8vF6vbf/kk0/ijTfeQEREBP7gD/4AN998s4XmEydOYMOGDbj66quRlJQEj2f0CwX09/fbc/nxnnvusQCusCxnYyTeXdqC5/ZX4aWDtRZmB885fWRoq2tnYOu1EUouO325o7As73jZ+XuecsIyQx9rls/G0exOnx/H6zqsbIIjpZlOOL/cdy1LHxju52XEYn5WHBKcoOp1gj47P3CEu8r5fvZXtFnZxumGLqv7ZckEvwf+LLqcm69/0Pk+B+2HxdHZC43QcoR+wyHn9e2qxIm6zt/q98yXy6/LAM/SkDTnxCDbCbWsPebfxRHxxs4+1DrB+HRjJw5WtmGHcyLz5vEGbDhSa5P12AqO3xNLY/Kcf4vleQm4vijNek+vn5tqKx7ydfL71G+7iIicjZlwy5YtqKysxMMPP2zZ8GIDseM2svzyyy/bqHJVVRX+4z/+AwUFBYFHfhu/hY985CP2Aj760Y/a58TR6WeeeQZf+cpX8Id/+Id44IEHMHfuXHvsYjSyLKPBwPhPG45Z5wSOKF8IR5M58vqfH7/COitc6luJNcocTf7qU/stVF5o4RB+aY7ufmF9IT5/TeGYlhCwJKKypduWyX7jaD32OWGZo7dni/YM1RcvyUmwkefClOih0eekKFtR8HzfD/ckB6va8If/s9tOLBiAR8Jnc6Idl5X+4vWzUOdszyWoOWLM0M4Jeizl4GIiZ59PMGBnJ0ZiVX4S1s5JsRF4lsm4wsd/sqKIiEx/U6oM4z//8z/tm8jPz7cR4erqaguvGRkZNrq8bt06ZGVlWUBuaGjAJz7xCcycORNf/vKXsWzZMvsa/NY4Ov31r38dhYWF+NznPofVq1fbY2fjdgzWvA1yBCxwH38gX/va16x2WmFZzsWgzGD2b6+fwCuHa39rlPdcnJMW6Xbh6/cttq4KrP29FBwt5ajrP7x01EZqL/abx9Hfj6zMw+9eM9PqfMcK/95e/4ATRP1W5sDaZpZnMJzydrqhE+3OY6wxjvK44I0Ig9cVbh852p0a40FStMdGdjnSzlF2fhx0fn4cMf7vrSXWAeOir48jy3EeZMVHwudszxIOPo9lKRx97nE+MpjzBIWBfVZqtNWMc+VCLsQS6/z9nLTH0hX9WouIyGhMqTKMV155Ba+//jp8Pp+VTnCyHssq+E3u2LHDyi94Hz9y9PnXv/61heVbbrkFcXFx9jUYbFnOcfDgQXR0dGDFihXnHaFubW3F22+/baUcnEzIco5NmzbZjTXQnByoMgw5F8Myw907J5tQMooV74azHy/75yVHWVi8FDXtPuyraLER3dFgduckPPZOZinFWOGvAEdiGXI5wY41yvz6/LvY+YMT4+akxdr9DKsMwQzPDPsckWapBifWlTo/s5POyUaxE7CP1bTjUHW7jQjXt/e+97O6EL4+trtjiUtzV589h98Ty06KMmKxblYKbpiXhuuKUnHVrGSsyE+07y0/Odq+b9aSM3DrV1pEREbrcsowxu3aJbtZMOgyCC9YsMBKKD75yU/iuuuus/vefPNNHD58GAMDAzbizI/8hodrmIdxgl9kZKSFbo4cnw/vZ9eNU6dO4dixY+/dWO/c1tYW2Erkt3GEd3BU0e43WFd8ORdk+ByWYlwKbs6wOp5Y18xShmW5Cbh1YQY+viYPn7m6AJ+4Mt8WO+ECHncszrQ+zVfPTrFAzUmI7Jfc7x+woMsSilPOiUeVE6Iv5bvlZL8l2fG40QnFty/KsEl/XFzlY6vz8Nm1M/Ep5/u43/nzNXNS7aSBIZkBWUREZKKMa6FfTEyMBeWHHnrIJuixvOK2227DHXfcYWUZDMsc6R2+MUycG0IsYDiBhql/pFFhjlDz7/jOd75jo8u8PfHEE/jxj3+M66+/PrCVyPvx/cSwxtHTEd5a78NNmNPYe/lyVn3jQiBsBTfa+mNuxtFTrtQ3kTjqzNKHVQVJNmnu99fPwt/cvQD/+uGl+H/O7S/vmI8v3zAbv+OE2Y84ofaepVkWpNlejx0pRrvgB1/fYico/8nNc/HvDy/H//nQIufrzsFHVuXi2rmpNnrP1y8iIjKZxi0sR0VFWflDXl6edbgYxpFidsLgMDhLK1gnwppmbsPRaN53NtYbs4VcbGysjTyfD4M0n88RaNaenH27lN7MElpYk1uYyhZjkYj1uC7aOYE9l9nmjG3kLqU/8DB2fuDflxQdcdHRUT7KYM3AyNtUwNfPWmV21bh6VgpuW5SBB1fk4NNrC/AlJzz/5Z0L8Ec3zrGaZtYRXwhfH3/mbPG2LC/RvvYozyFEREQm1LiF5bS0NAvLrCcennRH/Hy4fzJLLhhyuV1iYqJtO9yTmbjt8H3DreZGwlFChuZzbyONRovwrcFRZY6gsu+v3Wf//228n2H3nmVZl919IcIVhgIn+N65ONNGiy/0d9Ha2cnWjYKT66YC/rwY8vkzi/G6rGabJw0Mx1z4g10q2CWEpRRcYvpir489lvmz5yQ9/ZqKiMhUNW5hmd0reOPCJFyZj2UX7Hpx8uRJ7N6928Jxbm6ujSxz1HjevHk2+e+tt95CRUUFmpqacOTIEezbt8/CNXs1D6/2JzJWGNKW5sTbinKcuMeuDueWEbBlXEFKNK6Zm2JBlz18Lyfb8TnpTtC+a2mWBWGOaLPu92wMowygLGlg/TD7I0+XIMlvk9/7XUszrcaYEwbPXbyFP1ueKKzIS8Qtzs98sfOzV1AWEZGpbNy6YbAMo7OzE6+99pqNELPsgstds1UH71u/fr3VE3MEmqO/LLfghDwui83nNjc3W1cLbs8R5d/7vd+z0WWOFo8W/04tSiIXw/CWFudFWqwXLd39NkGNZQRsExfrjUBuUpQTptNsotvinAQrGbhcHle4jcJydLp/4Iy1SOMqczZa63HZ97HECZCfuiofV85KHnF1u6mKPxu+PraE45Ig7HbB37nh18eJgfMzY/HxK/Owbk6K83MYuy4fIiIiF3M53TDGrc8yu1twhJjB+Fe/+pWVUvCvys7Oxo033oh7770XRUVF79Uzs15548aN+OUvf2nt3vhnTtxbu3at9WBeunSp1SSPFv8uLUoio8U2clxEg+HuVEOnrbTHnr8sNRju/sDJZh8kKJ+NfxeDck1bj/P3daGtp8+Wl2YwH1ohMNxGncdyMZKJxIVP+PrYRq64vsMWfGE5CRc6Yb/kaK8LHudnyRMFERGRiXI5fZbHLSwT270Nt3RjKQX/KnbIYPkFFyThN3g2lmmUlZWhpqbG2sExHHMRk1mzZtm2lzKqrLAsl4O/DQzMXBCDAZojzCzNOLdcYizwF48LcfDvY7h0Oe9vBnJ2zQiGtylfX79zUsDFT/qc18cSjOERZv0aiojIZJhyYXkyKSyLiIiIyNkuJyyP/XCZiIiIiEiQUFgWERERERmBwrKIiIiIyAgUlkVERERERqCwLCIiIiIyAoVlEREREZERKCyLiIiIiIxAYVlEREREZAQKyyIiIiIiI1BYFhEREREZgcKyiIiIiMgIFJZFREREREagsCwiIiIiMgKFZRERERGRESgsi4iIiIiMQGFZRERERGQECssiIiIiIiNQWBYRERERGYHCsoiIiIjICBSWRURERERGoLAsIiIiIjIChWURERERkREoLIuIiIiIjEBhWURERERkBArLIiIiIiIjUFgWERERERmBwrKIiIiIyAgUlkVERERERqCwLCIiIiIyAoVlEREREZERKCyLiIiIiIxAYVlEREREZAQKyyIh4Ixz6x8YxKn6TuwqbcZO51Zc14le/yDO8EERERE5L4VlkSDX1NmLHaeb8P23SvD9t0vwo62ldvv+26ed+05j66lGNHT0BrYWERGRsyksiwSxtp5+7ClvwS93V+J7W07hyV0VeOFADV50bk/tqsRjTljmfRxpbu3u0yiziIjIORSWRYLY4ao2/NIJxb9ywnJLdz8GBn+ThgedZNzm3Pfcvmo8sbMc+ytb7T4RERH5DYVlkSDEyDvoBOMXD9Y4IbjN/nw+w/cfqenAs3urh2qYA/eJiIiIwrJIUOrtH0BpUzeK6zvR1NUXuPf8GI5ZgnHC2ba0sQu+voGhB0RERERhWSQY9fkHUd7cZWUW7IJxMf0DZ9De04+K5m70OEFbREREhigsiwShgTNn0Nnrt4+jxXrlzj7nOWfVNYuIiIQ6hWWRIOQKC0NilNv5OCNwz8WFO9smRDrPCR/9c0RERIKdwrJIEPJGhGFOWixSYz32+cV4XGFIifE4z4lBtNsVuFdEREQUlkWCUES4E35j3bgiPxG5iVG40FgxH8tKiMSqgkSkxXngdoKziIiIDNFRUSRIhc2YgRvnp+HKwmQbNY4Im/G+0MzPeV9yjBsrnVB926IMC9kiIiLyGzoyigSxBZnx+MiqXHx0dS4So91WlzwsjDXKzn33r8jGx9bkYXF2wvseFxEREYVlkaDG8DszJRoPrMjBPz+0FP/7jvn48g2z8aXrZ+Mvbp+Hf3pgiROm8zAnPVZBWURE5DwUlkWCXLTHhQInMK+bnYLbF2finqVZuGdZln2+bk4KClNjEONsIyIiIr9NYVkkRHDkODPea6PIc51bdkKk1ShrPFlERGRkCssiIiIiIiNQWBYRERERGYHCsoiIiIjICBSWRURERERGoLAsIiIiIjIChWURERERkREoLIuIiIiIjEBhWURERERkBArLIiIiIiIjUFgWERERERmBwrKIiIiIyAhmnHEEPg8qfFldXV145JFH0NHRge9+97uIiopCWJjOD0RERERCUXd3N77+9a9j586dePbZZxEdHY3w8PDAo+cX1GG5vb0df/zHf4yysjJ89atfhdfrxYwZMwJbiIiIiEgo8fl8+NnPfoaqqio8/fTTiI2NhcvlCjx6fkEblgcHB9HQ0GAh+ec//7mdNSgoi4iIiIQuxl5mxKuuugqPP/44UlNT4Xa7A4+eX1CPLPf19eHdd9/FT3/6Uxtq/7u/+ztkZGRc9AxCQltvby+ef/55vPXWW/ae4VnnxS7RSGjiDpejFI8++igWLFiAhx56yK5giZxPf38/du3ahf/3//6f7VvmzJmDiIiIwKMiv8EMw33L97///fdKSnV1fGzwZ8vfxYSEBFx99dUWlC9Wohu0YZn40lpbWy0s/+u//qsFoMLCQu2c5IJ6enrwne98B08++SReeOEFJCYm6gRLzmtgYMAOZA888IDtdP/sz/7M5kaInA9PxDds2IDf+Z3fsePRypUrLzqiJaGJ+YW1tV/72tcsx3zve9/TvKsx5Pf77cQjMjJyVCcgQR2WiWdmP/nJT6yY+5VXXlFYlotiWP7Wt76FX/ziF/aeSUpKUliW8xoOy/fccw/Wrl2Lv/zLv1RYlhExLL/00kv45Cc/afuWVatWKSzLeQ2H5a985SsWln/4wx8qLE8i/dRFREREREagsCwiIiIiMoKgD8u8ZJGXl4frr7/eeunpEoZcDN8jLNdZt24dPB6PJlTIiPjeYInOmjVrUFRUpImgckHct6Snp+OWW26xuRA6HsmFcH+yaNEiLF++XB29Jln43zoCnwct1iinpaVh7ty5Vh+mN5xcCN8ffM9kZmZi1qxZFob0npGRMPBwljrfKwxCCkAyEu5HGHrYqordU1iDqn2LnM/w+4KZhQN+ubm5tm/R+2VyBP0EPxERERGRy6UhEBERERGRESgsi4iIiIiMQGFZRERERGQECssiIiIiIiNQWBYRERERGUHQd8M4ePAg3nrrLezYsQMdHR22dPHq1atx4403YubMmWrzJPbeeO2117Bt27bAPUPY4ik2NhZ/8Rd/Yb0uJfRwOevTp09j69atOHDgAGpqatDe3o77778fN9xwAwoKCgJbDuHjfC/xPdXQ0GAt5ebMmYP77rvP9jfs9S7Bq6WlBcePH8fmzZtRWlqKpqYmJCQk4DOf+QwWL16MuLg4266/vx9PPPGEHZsqKyvtvmHcftmyZfjyl79s7x8JXu+88w527txp7xnuL9imlO0n+e9/6623WotBtjElRjXuj9544w1735w4ccLeR1wTgOtIcF2A+Ph421bGXtAmxeGD3AsvvIBNmzbZG41vwq6uLmzZssV2VM3NzfD7/YFnSKiqr6/H4cOHceTIEWRnZ2P27Nnv3RhwIiMjA1tKqBkcHLTAU1VVZUGot7fXDlY8uPHk+2wVFRX2GPctbW1tFnrYE3XPnj34yU9+gqNHj2p/E+R8Pp+dMPG9wPdKWVkZ3nzzTdTW1qKvry+w1dDxiWGH+xyefJ27z8nIyFA/3RDAwTzuS4j/5lwPgvuZjRs34sc//jHq6upsH0R8b/G99Pzzz9v7hide3P7UqVN48cUX8corr1jOkfERtIuS8IzrpZdewssvv2xnZp/97Gdx5513IiUlxQ5aPKPj2VtycrLO3kMcD1onT5605u//8A//gNtvvx3XXXcdrr32Wqxdu9Z2SLoCEZp48GltbbV/fy46Mm/ePBs5XrFiBZYsWWIn4MPefvttPP300ygpKcGnPvUp3HPPPTaqzJPyZ555xhYV4NfQ6HLwYqDhgAwXG+EVzO7ubgszN910k12F4P3EkyaODvb09Ng+5mtf+5rtc3jjn3kli8clBebgxkEahl5epeJxhyuBcnDm2LFjeO6557B+/Xo7/nDEmSft3/rWt+xYxX0Pr1bwcb7f9u7da4OD/Do8jul4NfaC9ifKs/gNGzbYzokhmcuL8qydl09vu+02u8TOUWeOKooQd0jccfFSFkcFh2+8X0IT/+15YOJ+46GHHsKqVavsYHQuhmoe+Hgg+8QnPmHhiJfdGX6+9KUv2X6IZRw8UZfgxcvmPMlmkLnjjjvsEvmFMNQwFHOfM7zf4UeeUCkoB79Pf/rTVm7D/QSvKPBk+kMf+pDdOKrMEy2erPPkimH59ddfx8KFC/Hggw/aEthclfh3f/d3beCPgz67du1CZ2dn4KvLWArKsMwDF8MyR3i48zl3h8XRIL4pefamN5YQR4D27dtnOy/uiL7whS/g29/+NsrLy+1yqsiF8EDGE28GnJUrV743gsg/c6SISxtzBIjlHCLDWK/MEh1ehfjoRz+KP//zP7fL7CzN0CX10MTswisU3HfwpIkn53w/MDjzfuaZ/Pz8wNawAR6WD/IjT9h5tULGXlCGZZ6FsZ6QN77ZOKnvbDExMUhMTLQ6MgUh4fuDZ+Z33XWXBZ2ioiKrE2N92De+8Y33JlKIjIQ1ytzfcCSaB67h0Wce8FgGxnpE7mu4nQhHlFmWwUlZnGzOUUIGINY4//rXv8bPf/5zO4FXYA49rEfmwA3fHyzdYl5hAOYEQL4fOADIYDxs+IooT9Crq6vfVxsvYycowzInT/AMjB950PJ4PIFHhvDNxfu5M+I2Etq4Q+IB63Of+5zdWG/KUh2G6CeffNIubXHkUGQk3N/whIrlXTxonV0zODy6zH2NDmRCfJ8wID/wwAN2GZ37nY997GNW5sNQxP0OgzPfVxI6OHrMTiqsP+bgDU+gWKbDAUDmFe5XmF2GO2QMY6bhfdxmeEKgjK2gDMs8OA3fzodnZ7ypCF4oLy/PDlKckMO6MV4yZ6uvz3/+83aixW4G3HmJjIT7Eu5vhvctZ+OfGZQvtE+S0DIcljmhi5P5uN/hpFHuczhJi900eJLOelUJftw/sEzrl7/8pU0UZu37I488Yh+J+w3uY4b3L+fbx/CmTDN+gvIny7MsTpLgR55pnVuXzMuhvJ+lGOeeoYkQD2a8tMX6MO7EdPlcLoT9uDl6zNFltn46u0UcD2KsOeS+Rp0w5EJ4yZ0BiccvXlLXyHJoYEnoN7/5TevetXTp0veCMjMMcTSZ5RfclzC7nFs+yvcJb8w0w8+RsRWUYZlnV7x0kZOTYyGHk7TOxkvqPHNXGycZCXdKvGTOtl98P+mkSi6EJTtsQ8lLoOydOjzJhu8jHsSKi4utPOPsVnMi5+LJFsMQ3z88+eJJuwQ3dshhb3a2umUNO8sv2HKSoXf4ShRPxlmSwdDMSaFnTxTmYA4nF3POBOfb8H0jYy9oyzD4puJkLb6B3n33XbuMzuDDNyaL5xmYr776ajsTk9DGlbY4qYI7HI4ADr9PuGob7+NJl0JO6BqeMMz3BU++GYgZZvg5R5F55YrlOryUztDMxQG40ABbP7EjD3u6c5SQ76OLtRKT6Y3vDY768X3B9wuDLy+x8/3D+/ie4XuH2xw6dMhqVLkdH+f7hauIssUggxIDE0eaJXixIxdHk7lwGssBWYLDXu7E9wn3PTzhZgDmxGG+JzjhnO8TjkYzx/Bz3sf3CkelhzvxyNgK2kVJiIF5//799kZigOZBjW9Khme+ob74xS9aCNJli9DGcMNgzDN0HrgYnjnJgvdzpIcT/tgt49yJohL8eKBikOFBjSdQ3Je8+uqr7y1mxAMaR45ZssNRQL5/2N+dBy6+nxh82AqM7yO2JGRdvPY3wYtXozhBj/McuB/Zvn27vXcYdPge4PuFgZpXqtjxgtvwuMTn8ASLi9fwxJ1rAvzO7/yODeaoDjV4sW0gJ3My9LJtKfcjHLBhEObJE//tmWN44+fcFw0vu8/3ELf71a9+ZSPNrHln+0FuOzwiLWNnhnMwCNreNNxxcdlI7oC4xDXfhDzIcdUbHriuuuoqe8PpjRXavv/979vECl594IGLJ1Js9cVFJbjz4UcetPQ+CT0c2eHB6Z/+6Z9s5b7hUcL/f3v3btswEARQkG2I3bAAxozUA5WrHzWnIow5gA4EKTGghH4TOXDmw/5uzZOQNE/2Cj1Asu/7mCwrjDxT63U2yU5BbVLkiwcKZa9xdY7Oy9/cJyd9L9kUWSNlimwyqEky+fPAzfV6ne73+xjcPJ/PkavsKbt5WJZlWtd1/FxjdW7bto36RBlmJVTDfcQH8eV2u/2+BKrJ8g+fj8djPKimedeAHQ+ZOFdiUL7j1MUyApEdH52YgCRoKYRceQhOJa6Y7thh10wpjgQsBbPGyrcuBbGS1v8kQSl6JCZTnFcSmtspu4L2Ck2TffLL2oUponPjH3MkO/Gm3fdzUxxb3dJgyTevnAXrOOKKByRMFI9pswbMZPFyuUzzPI+zVX46NzcPYsU78pC44lZCbME5kavUNNZ6NO/OlM+f+r1uP7/n9MVykiRJ8lctQyVJkiQfVCwnSZIkH1QsJ0mSJB9ULCdJkiRvTdMPxrcDdZBTm3YAAAAASUVORK5CYII=)"
      ],
      "metadata": {
        "id": "eSJa83uwq6Rs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Multiple Regression"
      ],
      "metadata": {
        "id": "Bsb29UFj2vwb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Multiple regression is like linear regression, but with more than one independent value, meaning that we try to **predict a value based on two or more variables**.\n",
        "\n",
        "We have a **result variable** that **we want to predict** in **function** of **two or more known variables**."
      ],
      "metadata": {
        "id": "cqQpTPLo2y-k"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The multiple regression explains how two or more independent variables determine a dependent resulting variable.\n",
        "\n",
        "Those two or more independent variables can take part with different impact over the result variable. This is called the **coefficient factor**.\n",
        "\n",
        "With the independent variables' coefficients we can know what would happen over the dependent variable if we increse or decrease one of the independent variables.\n",
        "\n",
        "We can imagine the multiple regression model as a polynomical function with two or more variables multiplied by a factor.\n",
        "\n",
        "Something such as:\n",
        "dependent_variable = a·x + b·y + c·z ...\n",
        "\n",
        "Where a, b and c are numerical values."
      ],
      "metadata": {
        "id": "d7ghSgVvCtqQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Scale Features"
      ],
      "metadata": {
        "id": "lAZzfD399X3f"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "When your data has different values, and even different measurement units, it can be difficult to compare them. What is kilograms compared to meters? Or altitude compared to time?\n",
        "\n",
        "The answer to this problem is scaling. We can scale data into new values that are easier to compare."
      ],
      "metadata": {
        "id": "yiPXszql9au3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The standardization method uses this formula:\n",
        "\n",
        "z = (x - u) / s\n",
        "\n",
        "Where z is the new value, x is the original value, u is the mean and s is the standard deviation.\n",
        "\n",
        "If you take the weight column from the data set above, the first value is 790, and the scaled value will be:\n",
        "\n",
        "  (790 - 1292.23) / 238.74 = -2.1\n",
        "\n",
        "If you take the volume column from the data set above, the first value is 1.0, and the scaled value will be:\n",
        "\n",
        "  (1.0 - 1.61) / 0.38 = -1.59\n",
        "\n",
        "Now you can compare -2.1 with -1.59 instead of comparing 790 with 1.0.\n",
        "\n",
        "You do not have to do this manually, the Python **sklearn** module has a method called **StandardScaler()** which returns a Scaler object with methods for transforming data sets."
      ],
      "metadata": {
        "id": "lfc17ywo9oJQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- It is necessary to import: **from sklearn.preprocessing import StandardScaler**\n",
        "\n",
        "- And use: **scale = StandardScaler()**"
      ],
      "metadata": {
        "id": "IenALSdE-Aqy"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "For more detail, click [here](https://www.w3schools.com/python/python_ml_scale.asp)"
      ],
      "metadata": {
        "id": "Y1MYOh4v90Oo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "rErblYNRqOva"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Code examples"
      ],
      "metadata": {
        "id": "xf56O6BrbWgj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Import libraries"
      ],
      "metadata": {
        "id": "CCPiKFWfWSRd"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "tmTVCqCJRfg-"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn.linear_model import LinearRegression\n",
        "import pandas as pd\n",
        "from scipy import stats\n",
        "import random\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Diamonds example - Linear Regression"
      ],
      "metadata": {
        "id": "pjAK1WQOWA5i"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "For more info click [here](https://discovery.cs.illinois.edu/learn/Towards-Machine-Learning/Machine-Learning-Models-in-Python-with-sk-learn/)\n",
        "\n",
        "The variable **model** uses the column variables *carat* and *price* from the dataframe comming from the [diamonds file](https://waf.cs.illinois.edu/discovery/diamonds.csv)\n",
        "\n"
      ],
      "metadata": {
        "id": "cHa_GeDhXbQb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Get the data"
      ],
      "metadata": {
        "id": "z_07li2mfFH5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "diamonds_dataset = pd.read_csv(\"https://waf.cs.illinois.edu/discovery/diamonds.csv\")\n",
        "\n",
        "print(diamonds_dataset)"
      ],
      "metadata": {
        "id": "dTnVAIMxTvS_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Train the model"
      ],
      "metadata": {
        "id": "ncCRdUBTe1o5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "diamonds_model = LinearRegression().fit(diamonds_dataset[ ['carat'] ]\n",
        "                                        , diamonds_dataset['price'])"
      ],
      "metadata": {
        "id": "dHTT1oZWe2oC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Check the regression score"
      ],
      "metadata": {
        "id": "e2oBnrTlfW8j"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "diamonds_model.score(diamonds_dataset[ ['carat'] ]\n",
        "                     , diamonds_dataset['price'])"
      ],
      "metadata": {
        "id": "PxaJQb2RfZKp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Test the model"
      ],
      "metadata": {
        "id": "aZS5WSkNtpsk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "diamonds_test = pd.DataFrame({'carat' : np.arange(0.1, 5.1, 0.1)})\n",
        "\n",
        "# Add a new column called 'predicted_price' to the diamonds_test dataset\n",
        "diamonds_test['predicted_price'] = diamonds_model.predict(diamonds_test)\n"
      ],
      "metadata": {
        "id": "yttwNt66trxm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Print the calculated results"
      ],
      "metadata": {
        "id": "pmt4RSUGuksk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(diamonds_test)"
      ],
      "metadata": {
        "id": "FoZVr6C8ucyh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Plot the actual data vs calculated data"
      ],
      "metadata": {
        "id": "jYVtE7GeunWV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(diamonds_dataset['price']\n",
        "         , diamonds_dataset['carat'])\n",
        "\n",
        "# To filter the values from a pandas.DataFrame:\n",
        "#    dataframe[\n",
        "#              dataframe['column_name'] < > = != value\n",
        "#             ]\n",
        "#             ['column_name_to_show']\n",
        "plt.plot(diamonds_test[diamonds_test['carat'] < 5]['predicted_price']\n",
        "                      , diamonds_test[diamonds_test['carat'] < 5]['carat'])\n",
        "\n",
        "plt.title('Price of the diamond in function of its weight')\n",
        "plt.xlabel('Diamond price (eur)')\n",
        "plt.ylabel('Diamond weight (carat)')\n",
        "\n",
        "plt.grid()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "hDxgDTveuptc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Car speed example - Linear Regression"
      ],
      "metadata": {
        "id": "PQ6SlyfN0nw3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The values of the python list **years_old** represent the age of the cars; the values of **speed* represent the velocity of the cars.\n",
        "\n",
        "The point here is to find the **Linear Regression** which represents the relation between the variables **years_old** and **speed**.\n",
        "\n",
        "The linear regression is used to determine one variable in function of another one.\n",
        "\n",
        "The values will be placed in the x-y axis.\n",
        "\n",
        "**cars** variable will be a pandas.DataFrame build from a python dictionary with two *keys* were each one will have a python list as a value inluding multiple values inside.\n",
        "\n"
      ],
      "metadata": {
        "id": "CIMzasjBYGUW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "years_old = [5,7,8,7,2,17,2,9,4,11,12,9,6]\n",
        "speed = [99,86,87,88,111,86,103,87,94,78,77,85,86]\n",
        "cars = pd.DataFrame({\n",
        "                      \"years_old\" : years_old\n",
        "                     , \"speed\" : speed\n",
        "                      })"
      ],
      "metadata": {
        "id": "vLRGSGW8SUqx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(cars)"
      ],
      "metadata": {
        "id": "NMYtqKe8S6A5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Prepare a plot from the columns *years_old* and *speed* from the DataFrame **cars**."
      ],
      "metadata": {
        "id": "_NKND0VsZ98l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(cars['years_old'], cars['speed'])\n",
        "\n",
        "plt.title('Cars age vs car speed')\n",
        "plt.xlabel('car age (years)')\n",
        "plt.ylabel('car speed (km/h)')\n",
        "\n",
        "plt.grid()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "TPIKd49jVyZ5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Train the model with sklearn"
      ],
      "metadata": {
        "id": "JZy8FofNzl1b"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use the function **fit** from **sklearn.linear_model.LinearRegression** to train the model.\n",
        "\n",
        "In the current example, the variable **model_cars** will store the result of the trained model having the Linear Regression among *years_old* and  *speed*."
      ],
      "metadata": {
        "id": "Qs1PLOxbab5U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_cars = LinearRegression()\n",
        "model_cars = model_cars.fit(cars[ [\"years_old\"] ], cars[\"speed\"] )"
      ],
      "metadata": {
        "id": "tPFJbhJEabQQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Test the model"
      ],
      "metadata": {
        "id": "SnLqSkxxzrhq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use the function **predict** from **sklearn.linear_model.LinearRegression** to calculate the resulting **speed** for each value.\n",
        "\n",
        "In the current example, we create a new pandas.DataFrame with some values for the variable **years_old**.\n",
        "\n",
        "After that, we create a new column called **calc_speed** which will have the predicted speed value in function of the years of the car and responding to the calculated linear regression."
      ],
      "metadata": {
        "id": "wdDgO7z0jKqK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sample_data_cars = pd.DataFrame({\"years_old\" :\n",
        "                                  [1, 3, 4, 5, 6, 7, 8, 9, 10, 20]})\n",
        "sample_data_cars[\"calc_speed\"] = model_cars.predict(sample_data_cars)"
      ],
      "metadata": {
        "id": "GWlmm8Zqc4YZ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(sample_data_cars)"
      ],
      "metadata": {
        "id": "Sw4micwvhTDm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(sample_data_cars[\"years_old\"], sample_data_cars[\"calc_speed\"])\n",
        "plt.scatter(sample_data_cars[\"years_old\"], sample_data_cars[\"calc_speed\"])\n",
        "\n",
        "plt.title('Linear Regression age vs speed')\n",
        "plt.xlabel('car age (years)')\n",
        "plt.ylabel('car speed (km/h)')\n",
        "plt.grid()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "cF0nLcwHj2uS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(cars[\"years_old\"], cars[\"speed\"])\n",
        "plt.plot(sample_data_cars[\"years_old\"], sample_data_cars[\"calc_speed\"])\n",
        "\n",
        "plt.title('Actual values and Linear Regression')\n",
        "plt.xlabel('car age (years)')\n",
        "plt.ylabel('car speed (km/h)')\n",
        "plt.grid()\n",
        "\n",
        "plt.show()\n"
      ],
      "metadata": {
        "id": "gzH6Se2LkNXa"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Check the linear regression line formula"
      ],
      "metadata": {
        "id": "EMgMk-Y5x_0X"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Once we trained the model and obtained the linear regression, we can get the line formula as follows:"
      ],
      "metadata": {
        "id": "lAquYd8zy84l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "m = model_cars.coef_\n",
        "b = model_cars.intercept_\n",
        "\n",
        "print('f(x) = ', m ,'x +', b)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tsvBUb43zEcF",
        "outputId": "0010c0a3-4ced-4d16-d68b-53c4ea340e2a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "name 'model_cars' is not defined",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-8-976d3dd0dd6c>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_cars\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcoef_\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_cars\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mintercept_\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'f(x) = '\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mm\u001b[0m \u001b[0;34m,\u001b[0m\u001b[0;34m'x +'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'model_cars' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we've got the **formula** of the **linear regression line**.\n",
        "\n",
        "Now for instance we can check what would be the speed of a 7.5 years old car:"
      ],
      "metadata": {
        "id": "yA9MAskE1OBk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y = lambda x : -1.75128771 * x + 103.10596026490066\n",
        "\n",
        "print(y(7.5))"
      ],
      "metadata": {
        "id": "RUiICt211agm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "If the check the values over the previous graph, it looks great!\n",
        "\n",
        "We can also check a single value by using the **model_cars**:"
      ],
      "metadata": {
        "id": "Qiei4yM91umr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(model_cars.predict([[7.5]]))"
      ],
      "metadata": {
        "id": "fUOkEDLy17LW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Nevertheless, our model is not perfect, because if we test the speed for a 0 years old car, we get 103.10... value, because it is the value of the intercept of the line over the y axis when x equals 0."
      ],
      "metadata": {
        "id": "RtVtiyLe2VyE"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Rain example - Linear Regression"
      ],
      "metadata": {
        "id": "qeTSsv6y0ttR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#first_20_days = np.random.uniform(0.8, 1, 20)\n",
        "#last_11_days = np.random.uniform(0, 0.20, 11)\n",
        "\n",
        "# Get k elements within a range(min, max, step) with random.choices()\n",
        "first_20_days = random.choices(range(0, 10, 1), k = 20)\n",
        "last_11_days = random.choices(range(11, 120, 1), k = 11)\n",
        "\n",
        "rain_values = np.concatenate((first_20_days, last_11_days), axis=None)\n",
        "\n",
        "print(rain_values)\n",
        "\n",
        "days = range(1, 32, 1)\n",
        "\n",
        "print(rain_values.size)\n",
        "\n",
        "rain_month = pd.DataFrame({\"day\" : days, \"rain\" : rain_values})\n",
        "\n",
        "print(rain_month)"
      ],
      "metadata": {
        "id": "qUPrbdTK0vw2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Plot the previous series"
      ],
      "metadata": {
        "id": "F5QWPp-nlONn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(rain_month[\"day\"], rain_month[\"rain\"])\n",
        "plt.scatter(rain_month[\"day\"], rain_month[\"rain\"])\n",
        "\n",
        "plt.title('Poured rain per day')\n",
        "plt.xlabel('Day of the month')\n",
        "plt.ylabel('Rain (ml/m^2)')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "g35QCoKvk1iR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Train the rain model"
      ],
      "metadata": {
        "id": "uxw9VGQ2E4cI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_rain = LinearRegression()\n",
        "model_rain = model_rain.fit(rain_month[ [\"day\"] ], rain_month[\"rain\"] )"
      ],
      "metadata": {
        "id": "gmqVtOfeEsGj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Test the model"
      ],
      "metadata": {
        "id": "2J7syQ7necOk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate a new DF called days representing the 31 days of the month\n",
        "days = pd.DataFrame({\"day\" : range(1, 32, 1)})\n",
        "\n",
        "# Test our model for each day\n",
        "days[\"predict\"] = model_rain.predict(days)\n",
        "\n",
        "print(days)"
      ],
      "metadata": {
        "id": "GDFLLV6EE3pn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "As we can see at the plot below, there is no such a good correlation between the days of the month and the amount of poured rain (orange series).\n",
        "\n",
        "That is why the linear regression represented in blue doesn't have a great accuracy."
      ],
      "metadata": {
        "id": "LzB_z-FxlrUi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot the continous line for the actual data\n",
        "plt.plot(rain_month[\"day\"], rain_month[\"rain\"])\n",
        "\n",
        "# Plot the scatter dots for the actual data\n",
        "plt.scatter(rain_month[\"day\"], rain_month[\"rain\"])\n",
        "\n",
        "# Plot the continuos line for the prediction\n",
        "# Represents the Linear Regression line\n",
        "plt.plot(days[\"day\"], days[\"predict\"])\n",
        "\n",
        "plt.title('Poured rain per day with the linear regression')\n",
        "plt.xlabel('Day of the month')\n",
        "plt.ylabel('Rain (ml/m^2)')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "fZGqps9QlUii"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Linear Regression score"
      ],
      "metadata": {
        "id": "IoiHT-O2cXn-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**LinearRegression.score(x, y)** returns the **coefficient of determination** of the **prediction** (a float number between 0 and 1; being 1 a perfect determination and 0 a poor one)."
      ],
      "metadata": {
        "id": "KqVcm6U_nMBL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_rain.score(rain_month[ [\"day\"] ], rain_month[\"rain\"])"
      ],
      "metadata": {
        "id": "6VTl55Itm7PK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Polynominal Regression example"
      ],
      "metadata": {
        "id": "Cwq4qp601vop"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**If your data points** clearly **will not fit a linear regression**(a straight line through all data points), **it might be ideal for polynomial regression**.\n",
        "\n",
        "Polynomial regression, like linear regression, uses the relationship between the variables x and y to find the best way to draw a line through the data points but by using a curved line.\n",
        "\n",
        "For that purpose, we will use:\n",
        "- **numpy.poly1d(numpy.polyfit(x, y, polynomic degree))** to create a polynomial model\n",
        "- **numpy.linspace(start, end, count of value points)** to specify how to draw the polynomial model curve."
      ],
      "metadata": {
        "id": "R1GqzWTm2c9E"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Create sample data for a polynomic function of degree = 3:"
      ],
      "metadata": {
        "id": "1BbuYLXK6aJm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "x = range(1, 23, 1)\n",
        "print(x[21])\n",
        "y = [100,90,80,82,60,60,55,58,60,65,70,70,75,78,76,78,79,80,90,99,99,100]"
      ],
      "metadata": {
        "id": "MBDAyuiM6GB2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Show the created points:"
      ],
      "metadata": {
        "id": "Rgg5_rPf6iHh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(x, y)"
      ],
      "metadata": {
        "id": "zBd_7c1t6hZY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Train the polynomic model:"
      ],
      "metadata": {
        "id": "hSaKSo9l7r_g"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "mymodel = np.poly1d(np.polyfit(x, y, 3))\n",
        "\n",
        "myline = np.linspace(1, 22, 100)"
      ],
      "metadata": {
        "id": "dSo0Xngn7rnl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Show the results:"
      ],
      "metadata": {
        "id": "qcEMYiCh76GD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(x, y)\n",
        "plt.plot(myline, mymodel(myline))\n",
        "\n",
        "plt.title('Actual points and Polynomic regression')\n",
        "plt.grid()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "775tqJGF77q6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Check the model significancy by looking the r-squared.\n",
        "\n",
        "If the model is significant, means that we can trust his predictions, if not, it won't bring trustfull predictions.\n",
        "\n",
        "The r-squared value ranges from 0 to 1, where 0 means no relationship, and 1 means 100% related.\n",
        "\n",
        "Python and the Sklearn module will compute this value for you, all you have to do is feed it with the x and y arrays:"
      ],
      "metadata": {
        "id": "047LAmnb8Cj7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import r2_score\n",
        "\n",
        "r_squared = r2_score(y, mymodel(x))\n",
        "\n",
        "print(r_squared)"
      ],
      "metadata": {
        "id": "Yrg0W6ny8Ffm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Test the model with individual values:"
      ],
      "metadata": {
        "id": "IleN0bac8p5k"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predicted_value = mymodel(15)\n",
        "\n",
        "print(predicted_value)"
      ],
      "metadata": {
        "id": "NTne7Wvz8xDc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Car example - Multiple Regression"
      ],
      "metadata": {
        "id": "DKKu5mLP42Oh"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Get the data from csv"
      ],
      "metadata": {
        "id": "E1kCwavSEnrd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Take a look at the data set below, it contains some information about cars."
      ],
      "metadata": {
        "id": "Rj5S4qQq5Xkp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "cars_complete_dataset = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/cars.csv')"
      ],
      "metadata": {
        "id": "25eVHGrq8sX5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(cars_complete_dataset)"
      ],
      "metadata": {
        "id": "2EAlAQ6889PO"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can predict the CO2 emission of a car based on the size of the engine, but with multiple regression we can throw in more variables, like the weight of the car, to make the prediction more accurate."
      ],
      "metadata": {
        "id": "8FTuZnsk8wsL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Prepare the data for the model"
      ],
      "metadata": {
        "id": "b-jXQ5koEuub"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "From the dataset, get the subset of *independent variables* that we want to use to predict (columns 'Weight' and 'Volume') and the result variable or also called *dependent variable* (column 'CO2').\n",
        "\n",
        "**Note**: It is common to name the list of independent values with a upper case X, and the list of dependent values with a lower case y."
      ],
      "metadata": {
        "id": "2PoDk6zs9HAf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X = cars_complete_dataset[['Weight', 'Volume']]\n",
        "y = cars_complete_dataset['CO2']"
      ],
      "metadata": {
        "id": "O4kBwJBD-SwU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Train the model"
      ],
      "metadata": {
        "id": "Ys6lCp-vEzEW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Import the library **linear_model** from **sklearn**."
      ],
      "metadata": {
        "id": "9yrpGm4X_crN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn import linear_model as lm"
      ],
      "metadata": {
        "id": "gTgNvhBr_0YA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Create the model co2_model and train it with the fit function and passing the independent variables ['Weight', 'Volume'] and the dependent variable ['CO2']."
      ],
      "metadata": {
        "id": "ByE7mZ5l_3zx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "co2_model = lm.LinearRegression().fit(X, y)\n",
        "print(type(co2_model))"
      ],
      "metadata": {
        "id": "t_GaXtox_dPZ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Test the model"
      ],
      "metadata": {
        "id": "p5qSX40DE36N"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now, we can use this **co2_model** to predict the CO2 generation of other motors based on their values of car weight and engine volume."
      ],
      "metadata": {
        "id": "4jBjXa_8AOGf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "cybertruck_weight = 3104\n",
        "cybertruck_engine = 6000\n",
        "\n",
        "cybertruck_co2 = co2_model.predict([[cybertruck_weight, cybertruck_engine]])\n",
        "\n",
        "print(cybertruck_co2)"
      ],
      "metadata": {
        "id": "Ljux4LA2AiD-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Get the coefficient factor"
      ],
      "metadata": {
        "id": "t3ldcBfqE6iv"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Calculate the **coefficient factor** for each independent variables."
      ],
      "metadata": {
        "id": "CxC4mtz6Ec5p"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(co2_model.coef_)"
      ],
      "metadata": {
        "id": "qgQxLc2jEkG_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The result array represents the coefficient values of weight and volume.\n",
        "\n",
        "Weight: 0.00755095\n",
        "Volume: 0.00780526\n",
        "\n",
        "These values tell us that if the weight increase by 1kg, the CO2 emission increases by 0.00755095g.\n",
        "\n",
        "And if the engine size (Volume) increases by 1 cm3, the CO2 emission increases by 0.00780526 g."
      ],
      "metadata": {
        "id": "FcpCvg48FZAF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Scale its features"
      ],
      "metadata": {
        "id": "Bm37MRUw-ckP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn import linear_model\n",
        "\n",
        "scale = StandardScaler()\n",
        "\n",
        "# Scale the values inside X (independent variables weight and engine)\n",
        "scaledX = scale.fit_transform(X)"
      ],
      "metadata": {
        "id": "SdzKKMF9-kW3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Regenerate the model:"
      ],
      "metadata": {
        "id": "hWFg6gjf_wNF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "new_model = linear_model.LinearRegression().fit(scaledX, y)"
      ],
      "metadata": {
        "id": "wxgJhl4p_yl8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Scale a single value to be tested:"
      ],
      "metadata": {
        "id": "GbMnq81xAY_K"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "scaled = scale.transform([[2300, 1.3]])"
      ],
      "metadata": {
        "id": "rIEhOeiCAafy"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Test the scaled value:"
      ],
      "metadata": {
        "id": "Vr5kWPRDAi5w"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predictedCO2 = new_model.predict([scaled[0]])\n",
        "\n",
        "print(predictedCO2)"
      ],
      "metadata": {
        "id": "nAOikw-TAkiT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "---"
      ],
      "metadata": {
        "id": "iGNj8SEN4zMD"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# References"
      ],
      "metadata": {
        "id": "o9LMPSz7h5e8"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- [Machine Learning Models in Python with sk-learn](https://discovery.cs.illinois.edu/learn/Towards-Machine-Learning/Machine-Learning-Models-in-Python-with-sk-learn/)\n",
        "- [Machine Learning - Linear Regression](https://www.w3schools.com/python/python_ml_linear_regression.asp)\n",
        "- [Matplotlib axis labels](https://www.scaler.com/topics/matplotlib/matplotlib-axis-label/)\n",
        "- [Machine Learning - Multiple Regression](https://www.w3schools.com/python/python_ml_multiple_regression.asp)"
      ],
      "metadata": {
        "id": "9_99KMHZh73p"
      }
    }
  ]
 }
No results found