{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ANrDBX0_EHfl"
      },
      "source": [
        "**PhucLam_Lab05**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hetF_D-MkzDt"
      },
      "source": [
        "# Lab 05 - Bài tập"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "pNrMxetqERBN"
      },
      "outputs": [],
      "source": [
        "# Import các thư viện thông dụng\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Import các distribution packages từ thư viện scipy\n",
        "from scipy.stats import binom\n",
        "from scipy.stats import poisson\n",
        "from scipy.stats import norm\n",
        "from scipy.stats import t"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "X6koQ0DDEGFn"
      },
      "source": [
        "**Bài 01:** Một bài thi trắc nghiệm gồm 10 câu hỏi, mỗi câu có 4 phương án trả lời trong đó chỉ có một phương án đúng. Bạn hãy:\n",
        "\n",
        "a. Một sinh viên không học bài làm bài bằng cách chọn ngẫu nhiên một phương án cho mỗi câu hỏi. Bạn hãy dùng hàm random để in ra số câu đúng của bạn sinh viên trong 10 lần kiểm tra.\n",
        "\n",
        "b. Giả sử mỗi câu đúng được 4 điểm, mỗi câu sai bị trừ 2 điểm. Tính xác suất để sinh viên này được 4 điểm\n",
        "\n",
        "c. Gọi X là số câu trả lời đúng. Tính E(X) và Var(X)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4JswIH66EGFn",
        "outputId": "06a9d13e-6a71-4cda-c895-64794b68291f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "25\n"
          ]
        }
      ],
      "source": [
        "# a. Một sinh viên không học bài làm bài bằng cách chọn ngẫu nhiên một phương án cho mỗi câu hỏi\n",
        "import random\n",
        "\n",
        "def simulate_tests(num_tests=10):\n",
        "    correct_count = 0\n",
        "    for _ in range(num_tests):\n",
        "        correct_answers = 0\n",
        "        for _ in range(10):  # 10 câu hỏi\n",
        "            if random.random() < 0.25:  # Xác suất chọn đúng\n",
        "                correct_answers += 1\n",
        "        correct_count += correct_answers\n",
        "    return correct_count\n",
        "\n",
        "# In ra số câu đúng của sinh viên trong 10 lần kiểm tra\n",
        "print(simulate_tests())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hus8aJ-DHquE",
        "outputId": "645e4442-2d4b-44df-830c-e5c264f4f489"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Xác suất để sinh viên này được 4 điểm là: 0.2503\n"
          ]
        }
      ],
      "source": [
        "from scipy.stats import binom\n",
        "\n",
        "# Số câu hỏi\n",
        "n = 10\n",
        "# Xác suất đúng cho mỗi câu (1/4)\n",
        "p = 0.25\n",
        "\n",
        "# Xác suất để có 3 câu đúng (k = 3)\n",
        "k = 3\n",
        "prob_4_points = binom.pmf(k, n, p)\n",
        "print(f\"Xác suất để sinh viên này được 4 điểm là: {prob_4_points:.4f}\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "b5YvMkUkHobc",
        "outputId": "6ae27ea4-73bc-494e-93d6-b74ec150081d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "E(X) = 2.50\n",
            "Var(X) = 1.88\n"
          ]
        }
      ],
      "source": [
        "# Tính E(X) và Var(X)\n",
        "E_X = n * p\n",
        "Var_X = n * p * (1 - p)\n",
        "\n",
        "print(f\"E(X) = {E_X:.2f}\")\n",
        "print(f\"Var(X) = {Var_X:.2f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5_kbSiYkEGFn"
      },
      "source": [
        "**Bài 02:** Một trung tâm bưu điện nhận được 3 cuộc gọi điện thoại mỗi phút. Tính xác suất để trung tâm nhận được 1 cuộc gọi, 2 cuộc gọi, 3 cuộc gọi trong vòng một phút biết số cuộc gọi trong một phút có phân phối Poisson."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_3SZ4PysEGFn",
        "outputId": "2a6f8f84-8ef1-4ccf-b774-c9dced3b2ffa"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Xác suất nhận được 1 cuộc gọi: 0.1494\n",
            "Xác suất nhận được 2 cuộc gọi: 0.2240\n",
            "Xác suất nhận được 3 cuộc gọi: 0.2240\n"
          ]
        }
      ],
      "source": [
        "import math\n",
        "\n",
        "# Hàm tính xác suất theo phân phối Poisson\n",
        "def poisson_probability(lmbda, k):\n",
        "    return (lmbda ** k) * math.exp(-lmbda) / math.factorial(k)\n",
        "\n",
        "# Tham số lambda (số cuộc gọi trung bình mỗi phút)\n",
        "lmbda = 3\n",
        "\n",
        "# Tính xác suất cho 1, 2, và 3 cuộc gọi\n",
        "p_1_call = poisson_probability(lmbda, 1)\n",
        "p_2_calls = poisson_probability(lmbda, 2)\n",
        "p_3_calls = poisson_probability(lmbda, 3)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"Xác suất nhận được 1 cuộc gọi: {p_1_call:.4f}\")\n",
        "print(f\"Xác suất nhận được 2 cuộc gọi: {p_2_calls:.4f}\")\n",
        "print(f\"Xác suất nhận được 3 cuộc gọi: {p_3_calls:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "unuJFn76EGFn"
      },
      "source": [
        "**Bài 03:** Trọng lượng (đơn vị gam) của một loại trái cây có phân phối chuẩn với µ = 500 (gam) gam và $σ^2 = 16 (gam^2)$. Trái cây thu hoạch được phân loại theo trọng lượng như sau:\n",
        "\n",
        "Loại 1: trên 505 gam\n",
        "\n",
        "Loại 2: từ 495 – 505 gam\n",
        "\n",
        "Loại 3: dười 495 gam\n",
        "\n",
        "a. Hãy tính tỷ lệ của mỗi loại.\n",
        "b. Bạn hãy mô phỏng lấy mẫu  20 trái cây trên. Bạn hãy vẽ histogram trọng lượng trái cây và tính giá trị trung bình và độ lệch chuẩn trái cây từ mẫu\n",
        "c. Bạn hãy mô phỏng lấy 100 trái cây trên. Bạn hãy vẽ histogram trọng lượng trái cây và tính giá trị trung bình và độ lệch chuẩn trái cây từ mẫu\n",
        "d. Từ hai cách lấy mẫu ở câu b, c bạn có nhận xét gì không?"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hpPHDaHcEGFo",
        "outputId": "897b9d1b-f550-4224-f35e-e6a49124c6c6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Tỷ lệ loại 1 (trên 505 gam): 0.1056\n",
            "Tỷ lệ loại 2 (từ 495 đến 505 gam): 0.7887\n",
            "Tỷ lệ loại 3 (dưới 495 gam): 0.1056\n"
          ]
        }
      ],
      "source": [
        "import scipy.stats as stats\n",
        "\n",
        "#a. Hãy tính tỷ lệ của mỗi loại\n",
        "# Tham số phân phối\n",
        "mu = 500\n",
        "sigma = 4\n",
        "\n",
        "# Tính xác suất cho từng loại\n",
        "p_loai_1 = 1 - stats.norm.cdf(505, mu, sigma)\n",
        "p_loai_2 = stats.norm.cdf(505, mu, sigma) - stats.norm.cdf(495, mu, sigma)\n",
        "p_loai_3 = stats.norm.cdf(495, mu, sigma)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"Tỷ lệ loại 1 (trên 505 gam): {p_loai_1:.4f}\")\n",
        "print(f\"Tỷ lệ loại 2 (từ 495 đến 505 gam): {p_loai_2:.4f}\")\n",
        "print(f\"Tỷ lệ loại 3 (dưới 495 gam): {p_loai_3:.4f}\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 998
        },
        "id": "mwqJEDGdKN3y",
        "outputId": "a51c0393-6540-4c60-85bf-ecceb43b9d67"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Trung bình (20 trái cây): 500.82\n",
            "Độ lệch chuẩn (20 trái cây): 3.67\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Trung bình (100 trái cây): 500.28\n",
            "Độ lệch chuẩn (100 trái cây): 4.44\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "#b. Bạn hãy mô phỏng lấy mẫu 20 trái cây trên\n",
        "# Mô phỏng lấy mẫu 20 trái cây\n",
        "samples_20 = np.random.normal(mu, sigma, 20)\n",
        "\n",
        "# Vẽ histogram cho 20 trái cây\n",
        "plt.hist(samples_20, bins=10, alpha=0.7, color='blue', edgecolor='black')\n",
        "plt.title('Histogram của 20 trái cây')\n",
        "plt.xlabel('Trọng lượng (gam)')\n",
        "plt.ylabel('Tần suất')\n",
        "plt.show()\n",
        "\n",
        "# Tính trung bình và độ lệch chuẩn của 20 trái cây\n",
        "mean_20 = np.mean(samples_20)\n",
        "std_20 = np.std(samples_20, ddof=1)\n",
        "print(f\"Trung bình (20 trái cây): {mean_20:.2f}\")\n",
        "print(f\"Độ lệch chuẩn (20 trái cây): {std_20:.2f}\")\n",
        "\n",
        "# Mô phỏng lấy mẫu 100 trái cây\n",
        "samples_100 = np.random.normal(mu, sigma, 100)\n",
        "\n",
        "# Vẽ histogram cho 100 trái cây\n",
        "plt.hist(samples_100, bins=10, alpha=0.7, color='green', edgecolor='black')\n",
        "plt.title('Histogram của 100 trái cây')\n",
        "plt.xlabel('Trọng lượng (gam)')\n",
        "plt.ylabel('Tần suất')\n",
        "plt.show()\n",
        "\n",
        "# Tính trung bình và độ lệch chuẩn của 100 trái cây\n",
        "mean_100 = np.mean(samples_100)\n",
        "std_100 = np.std(samples_100, ddof=1)\n",
        "print(f\"Trung bình (100 trái cây): {mean_100:.2f}\")\n",
        "print(f\"Độ lệch chuẩn (100 trái cây): {std_100:.2f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XP3ERxAXEGFo"
      },
      "source": [
        "**Bài 04:** Mô phỏng tung một con xúc sắc cân đối đồng chất 5000 lần. Dựa vào giá trị mô phỏng, bạn hãy tìm các giá trị xác suất dưới đây:\n",
        "\n",
        "a. Xác suất để  số chấm xuất hiện là 4\n",
        "\n",
        "b. Xác suất để số chấm xuất hiện lớn hơn hoặc bằng 4\n",
        "\n",
        "c. Giả sử biết số chấm xuất hiện lớn hơn hoặc bằng 4. Hãy tìm xác suất để mặt 6 chấm xuất hiện."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SqQrzOqDEGFo",
        "outputId": "eb930f7f-e2fd-4078-cd60-fdcb258673d5",
        "vscode": {
          "languageId": "plaintext"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Xác suất để số chấm xuất hiện là 4: 0.1684\n",
            "Xác suất để số chấm xuất hiện lớn hơn hoặc bằng 4: 0.5030\n",
            "Xác suất để mặt 6 chấm xuất hiện, biết rằng số chấm >= 4: 0.3264\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "\n",
        "# Mô phỏng tung xúc xắc 5000 lần\n",
        "np.random.seed(42)  # Đặt seed để tái tạo kết quả\n",
        "dice_rolls = np.random.randint(1, 7, 5000)  # Tung xúc xắc, các giá trị từ 1 đến 6\n",
        "\n",
        "# Phần a: Xác suất để số chấm xuất hiện là 4\n",
        "count_4 = np.sum(dice_rolls == 4)\n",
        "p_4 = count_4 / 5000\n",
        "print(f\"Xác suất để số chấm xuất hiện là 4: {p_4:.4f}\")\n",
        "\n",
        "# Phần b: Xác suất để số chấm xuất hiện lớn hơn hoặc bằng 4\n",
        "count_gte_4 = np.sum(dice_rolls >= 4)\n",
        "p_gte_4 = count_gte_4 / 5000\n",
        "print(f\"Xác suất để số chấm xuất hiện lớn hơn hoặc bằng 4: {p_gte_4:.4f}\")\n",
        "\n",
        "# Phần c: Xác suất để mặt 6 chấm xuất hiện, biết rằng số chấm lớn hơn hoặc bằng 4\n",
        "count_6_given_gte_4 = np.sum(dice_rolls[dice_rolls >= 4] == 6)\n",
        "p_6_given_gte_4 = count_6_given_gte_4 / count_gte_4\n",
        "print(f\"Xác suất để mặt 6 chấm xuất hiện, biết rằng số chấm >= 4: {p_6_given_gte_4:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aqsKmkdeEGFo"
      },
      "source": [
        "**Bài 05:** Tạo 5000 số ngẫu nhiên có phân phối nhị thức với n=50, p=0.7 Bạn Hãy\n",
        "\n",
        "a. Tìm các giá trị thống kê: min, max, Q1, Q2, Q3\n",
        "\n",
        "b. Vẽ đồ thị boxplot\n",
        "\n",
        "c. Vẽ các đồ thị: tần số, tần suất"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "VUDpu98GEGFo",
        "outputId": "41600990-93f0-4d1c-caab-b3e90d767d52",
        "vscode": {
          "languageId": "plaintext"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Min: 24\n",
            "Max: 47\n",
            "Q1 (25th percentile): 33.0\n",
            "Q2 (Median): 35.0\n",
            "Q3 (75th percentile): 37.0\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Phần a: Tạo 5000 số ngẫu nhiên với phân phối nhị thức\n",
        "n = 50\n",
        "p = 0.7\n",
        "samples = np.random.binomial(n, p, 5000)\n",
        "\n",
        "# Tính các giá trị thống kê: min, max, Q1, Q2, Q3\n",
        "min_value = np.min(samples)\n",
        "max_value = np.max(samples)\n",
        "Q1 = np.percentile(samples, 25)\n",
        "Q2 = np.median(samples)\n",
        "Q3 = np.percentile(samples, 75)\n",
        "\n",
        "# In các giá trị thống kê\n",
        "print(f\"Min: {min_value}\")\n",
        "print(f\"Max: {max_value}\")\n",
        "print(f\"Q1 (25th percentile): {Q1}\")\n",
        "print(f\"Q2 (Median): {Q2}\")\n",
        "print(f\"Q3 (75th percentile): {Q3}\")\n",
        "\n",
        "# Phần b: Vẽ boxplot\n",
        "plt.boxplot(samples, vert=False)\n",
        "plt.title(\"Boxplot của mẫu phân phối nhị thức\")\n",
        "plt.xlabel(\"Giá trị\")\n",
        "plt.show()\n",
        "\n",
        "# Phần c: Vẽ biểu đồ tần số (Histogram)\n",
        "plt.hist(samples, bins=15, edgecolor='black', alpha=0.7)\n",
        "plt.title(\"Biểu đồ tần số (Histogram)\")\n",
        "plt.xlabel(\"Giá trị\")\n",
        "plt.ylabel(\"Tần số\")\n",
        "plt.show()\n",
        "\n",
        "# Vẽ biểu đồ tần suất (Biểu đồ cột của tỷ lệ)\n",
        "unique, counts = np.unique(samples, return_counts=True)\n",
        "relative_frequencies = counts / len(samples)\n",
        "\n",
        "plt.bar(unique, relative_frequencies, edgecolor='black', alpha=0.7)\n",
        "plt.title(\"Biểu đồ tần suất\")\n",
        "plt.xlabel(\"Giá trị\")\n",
        "plt.ylabel(\"Tần suất\")\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aqWfhZoHEGFo"
      },
      "source": [
        "**Bài 6**\n",
        "\n",
        "Giả sử X là biến ngẫu nhiên rời rạc\n",
        "\n",
        "a. Hãy cho biết ý nghĩa câu lệnh sau: `binom.pmf(4, 10, 0.6)`\n",
        "\n",
        "b. Giả sử X có phân phối nhị thức `X ~ B(10, 0.6)`. Hãy tính các giá trị sau:  \n",
        "$P(X \\leq 5), P(X \\le 5), P(X \\ge 4), P(X = 5) $\n",
        "\n",
        "c. Hãy cho biết ý nghĩa câu lệnh sau: `poisson.pmf(4, 3)`\n",
        "\n",
        "d. Giả sử X có phân phối Poisson `X ~ P(3)`. Hãy tính các giá trị sau:  \n",
        "$P(X \\leq 5), P(X \\le 5), P(X \\ge 4), P(X = 5) $"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7aTGQI2iLd80"
      },
      "source": [
        "Phần a: Ý nghĩa của câu lệnh binom.pmf(4, 10, 0.6)\n",
        "binom.pmf(4, 10, 0.6): Đây là hàm khối xác suất (Probability Mass Function - PMF) của phân phối nhị thức (binomial distribution). Hàm này tính xác suất để biến ngẫu nhiên\n",
        "𝑋\n",
        "X nhận giá trị cụ thể, ở đây là 4, trong một phân phối nhị thức có các tham số:\n",
        "𝑛\n",
        "=\n",
        "10\n",
        "n=10 (số lần thử),\n",
        "𝑝\n",
        "=\n",
        "0.6\n",
        "p=0.6 (xác suất thành công mỗi lần thử).\n",
        "Cụ thể, câu lệnh này tính xác suất để có đúng 4 lần thành công trong 10 lần thử, khi xác suất thành công mỗi lần thử là 0.6."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FIer-tGjLhST",
        "outputId": "881b6a24-93aa-4dc7-f55c-fc4dad492ad6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "P(X <= 5): 0.3669\n",
            "P(X >= 4): 0.9452\n",
            "P(X = 5): 0.2007\n"
          ]
        }
      ],
      "source": [
        "import scipy.stats as stats\n",
        "\n",
        "# Tham số của phân phối nhị thức\n",
        "n = 10  # Số lần thử\n",
        "p = 0.6  # Xác suất thành công mỗi lần thử\n",
        "\n",
        "# Tính các xác suất\n",
        "P_X_le_5 = stats.binom.cdf(5, n, p)  # P(X <= 5)\n",
        "P_X_ge_4 = 1 - stats.binom.cdf(3, n, p)  # P(X >= 4)\n",
        "P_X_eq_5 = stats.binom.pmf(5, n, p)  # P(X = 5)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"P(X <= 5): {P_X_le_5:.4f}\")\n",
        "print(f\"P(X >= 4): {P_X_ge_4:.4f}\")\n",
        "print(f\"P(X = 5): {P_X_eq_5:.4f}\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UzONlW0sLnOL",
        "outputId": "074aa9da-6767-46ef-b477-854eca3ed2f4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "P(X <= 5) - Poisson: 0.9161\n",
            "P(X >= 4) - Poisson: 0.3528\n",
            "P(X = 5) - Poisson: 0.1008\n"
          ]
        }
      ],
      "source": [
        "# Tham số của phân phối Poisson\n",
        "lambda_poisson = 3  # Số sự kiện trung bình\n",
        "\n",
        "# Tính các xác suất\n",
        "P_X_le_5_poisson = stats.poisson.cdf(5, lambda_poisson)  # P(X <= 5)\n",
        "P_X_ge_4_poisson = 1 - stats.poisson.cdf(3, lambda_poisson)  # P(X >= 4)\n",
        "P_X_eq_5_poisson = stats.poisson.pmf(5, lambda_poisson)  # P(X = 5)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"P(X <= 5) - Poisson: {P_X_le_5_poisson:.4f}\")\n",
        "print(f\"P(X >= 4) - Poisson: {P_X_ge_4_poisson:.4f}\")\n",
        "print(f\"P(X = 5) - Poisson: {P_X_eq_5_poisson:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T2qyX1esEGFp"
      },
      "source": [
        "**Bài 7**\n",
        "\n",
        "a. Cho biết ý nghĩa của các câu lệnh sau:\n",
        "\n",
        "- `norm.cdf(2)`\n",
        "\n",
        "- `norm.cdf(2, 1, 1)`\n",
        "\n",
        "- `norm.cdf(2, 1, 2)`\n",
        "\n",
        "b. Cho biết kết quả của câu lệnh sau: `norm.ppf(norm.cdf(2))`\n",
        "\n",
        "c. Tính các giá trị sau và vẽ hình minh họa cho các giá trị tính được:\n",
        "\n",
        "- `norm.ppf(0.975)`\n",
        "\n",
        "- `norm.ppf(0.975, 1, 1)`\n",
        "\n",
        "- `norm.ppf(0.975, 1, 2)`\n",
        "\n",
        "d. Giả sử biến ngẫu nhiên X có phân phối chuẩn với kỳ vọng là 24 và phương sai là 16. Hãy tính các giá trị sau:\n",
        "\n",
        "- $P(X \\leq 20)$\n",
        "- $P(X \\ge 29.5)$\n",
        "- $P(X = 23.8)$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NUYIfdFbL4i9"
      },
      "source": [
        "Phần a: Ý nghĩa của các câu lệnh\n",
        "norm.cdf(2):\n",
        "\n",
        "Tính giá trị của hàm phân phối tích lũy (CDF) của phân phối chuẩn tắc (standard normal distribution) tại\n",
        "𝑥\n",
        "=\n",
        "2\n",
        "x=2.\n",
        "Điều này có nghĩa là tính xác suất để biến ngẫu nhiên chuẩn tắc\n",
        "𝑍\n",
        "≤\n",
        "2\n",
        "Z≤2, tức là\n",
        "𝑃\n",
        "(\n",
        "𝑍\n",
        "≤\n",
        "2\n",
        ")\n",
        "P(Z≤2), với\n",
        "𝑍\n",
        "∼\n",
        "𝑁\n",
        "(\n",
        "0\n",
        ",\n",
        "1\n",
        ")\n",
        "Z∼N(0,1).\n",
        "norm.cdf(2, 1, 1):\n",
        "\n",
        "Tính giá trị của hàm phân phối tích lũy của phân phối chuẩn có kỳ vọng (mean)\n",
        "𝜇\n",
        "=\n",
        "1\n",
        "μ=1 và độ lệch chuẩn (standard deviation)\n",
        "𝜎\n",
        "=\n",
        "1\n",
        "σ=1 tại\n",
        "𝑥\n",
        "=\n",
        "2\n",
        "x=2.\n",
        "Điều này có nghĩa là tính xác suất để biến ngẫu nhiên\n",
        "𝑋\n",
        "≤\n",
        "2\n",
        "X≤2, tức là\n",
        "𝑃\n",
        "(\n",
        "𝑋\n",
        "≤\n",
        "2\n",
        ")\n",
        "P(X≤2), với\n",
        "𝑋\n",
        "∼\n",
        "𝑁\n",
        "(\n",
        "1\n",
        ",\n",
        "1\n",
        ")\n",
        "X∼N(1,1).\n",
        "norm.cdf(2, 1, 2):\n",
        "\n",
        "Tương tự như trên, nhưng lần này độ lệch chuẩn là\n",
        "𝜎\n",
        "=\n",
        "2\n",
        "σ=2.\n",
        "Tính xác suất để\n",
        "𝑋\n",
        "≤\n",
        "2\n",
        "X≤2, tức là\n",
        "𝑃\n",
        "(\n",
        "𝑋\n",
        "≤\n",
        "2\n",
        ")\n",
        "P(X≤2), với\n",
        "𝑋\n",
        "∼\n",
        "𝑁\n",
        "(\n",
        "1\n",
        ",\n",
        "2\n",
        ")\n",
        "X∼N(1,2)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Mmwjz8nMEGFp",
        "outputId": "79d8dbe2-72cb-4435-8e44-92b949bf24e5"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Kết quả của norm.ppf(norm.cdf(2)): 2.0000000000000004\n"
          ]
        }
      ],
      "source": [
        "import scipy.stats as stats\n",
        "\n",
        "result_b = stats.norm.ppf(stats.norm.cdf(2))\n",
        "print(f\"Kết quả của norm.ppf(norm.cdf(2)): {result_b}\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 505
        },
        "id": "So9KMa3WL8y7",
        "outputId": "e3a4d18b-bfee-4cf7-f70c-d0a08443b83a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "norm.ppf(0.975): 1.959963984540054\n",
            "norm.ppf(0.975, 1, 1): 2.959963984540054\n",
            "norm.ppf(0.975, 1, 2): 4.919927969080108\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Tính các giá trị\n",
        "value_1 = stats.norm.ppf(0.975)\n",
        "value_2 = stats.norm.ppf(0.975, 1, 1)\n",
        "value_3 = stats.norm.ppf(0.975, 1, 2)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"norm.ppf(0.975): {value_1}\")\n",
        "print(f\"norm.ppf(0.975, 1, 1): {value_2}\")\n",
        "print(f\"norm.ppf(0.975, 1, 2): {value_3}\")\n",
        "\n",
        "# Vẽ hình minh họa\n",
        "x = np.linspace(-4, 4, 1000)\n",
        "y = stats.norm.pdf(x)\n",
        "\n",
        "plt.plot(x, y, label=\"N(0, 1)\")\n",
        "plt.axvline(value_1, color='r', linestyle='--', label=f\"norm.ppf(0.975) = {value_1:.2f}\")\n",
        "plt.title(\"Hàm mật độ xác suất N(0, 1)\")\n",
        "plt.legend()\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F5jgs5L0NWPN",
        "outputId": "4a202ca8-c1b6-416a-ac99-e590829d9d1e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "P(X <= 20): 0.1587\n",
            "P(X >= 29.5): 0.0846\n",
            "P(X = 23.8): 0 (Vì xác suất tại một điểm trong phân phối liên tục là 0)\n"
          ]
        }
      ],
      "source": [
        "# Tham số của phân phối chuẩn\n",
        "mu = 24\n",
        "sigma = 4\n",
        "\n",
        "# Tính xác suất\n",
        "P_X_le_20 = stats.norm.cdf(20, mu, sigma)\n",
        "P_X_ge_29_5 = 1 - stats.norm.cdf(29.5, mu, sigma)\n",
        "\n",
        "# In kết quả\n",
        "print(f\"P(X <= 20): {P_X_le_20:.4f}\")\n",
        "print(f\"P(X >= 29.5): {P_X_ge_29_5:.4f}\")\n",
        "print(f\"P(X = 23.8): 0 (Vì xác suất tại một điểm trong phân phối liên tục là 0)\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z7N86s_VEGFp"
      },
      "source": [
        "**Bài 8**\n",
        "\n",
        "Một công ty sản xuất làm ra các chi tiết máy với độ dài các chi tiết có phân phối chuẩn với kỳ vọng là 3000 mm và độ lệch chuẩn là 3 mm. Yêu cầu cho các chi tiết máy này là phải có chiều dài từ 2993 mm đến 3007 mm. Tính tỷ lệ sản phẩm lỗi của công ty"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UvHL7z-aEGFp",
        "outputId": "d7dcbf9e-c3ff-4045-ee9a-e38505cf6c3e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Xác suất để chi tiết máy nằm trong khoảng 2993 mm đến 3007 mm: 0.9804\n",
            "Tỷ lệ sản phẩm lỗi của công ty: 0.0196\n"
          ]
        }
      ],
      "source": [
        "import scipy.stats as stats\n",
        "\n",
        "# Tham số của phân phối chuẩn\n",
        "mu = 3000\n",
        "sigma = 3\n",
        "\n",
        "# Giá trị z cho ngưỡng dưới và ngưỡng trên\n",
        "z1 = (2993 - mu) / sigma\n",
        "z2 = (3007 - mu) / sigma\n",
        "\n",
        "# Tính xác suất P(2993 ≤ X ≤ 3007)\n",
        "P_within_range = stats.norm.cdf(z2) - stats.norm.cdf(z1)\n",
        "\n",
        "# Tỷ lệ sản phẩm lỗi\n",
        "P_outside_range = 1 - P_within_range\n",
        "\n",
        "# In kết quả\n",
        "print(f\"Xác suất để chi tiết máy nằm trong khoảng 2993 mm đến 3007 mm: {P_within_range:.4f}\")\n",
        "print(f\"Tỷ lệ sản phẩm lỗi của công ty: {P_outside_range:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WsJOjJMkEGFp"
      },
      "source": [
        "**Bài 9**\n",
        "\n",
        "Giả sử điểm thi TOEIC là một biến ngẫu nhiên có phân phối chuẩn với kỳ vọng là 500, độ lệch chuẩn là 8. Bạn hãy:\n",
        "\n",
        "a. Mô phỏng bằng cách phát sinh ngẫu nhiên điểm thi TOIEC của một nhóm gồm 50 người.\n",
        "\n",
        "b. Tính điểm trung bình và độ lệch chuẩn về điểm của nhóm.\n",
        "\n",
        "c. Tìm miền giá trị, và miền phân vị (IQR) của nhóm\n",
        "\n",
        "d. Cho biết tỷ lệ đạt trên 450 điểm của nhóm.\n",
        "\n",
        "e. Vẽ đồ thị histogram của nhóm và so sánh với đồ thị hàm mật độ của phân phối chuẩn với kỳ vọng là 500 và độ lệch chuẩn là 8.\n",
        "\n",
        "f. Thay vì mô phỏng 50 người, hãy mô phỏng nhóm với 100 người thi. Bạn có nhận xét gì?"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 722
        },
        "id": "--5HP07nEGFp",
        "outputId": "11dc5d81-ecb4-4694-d2d3-fa91d13dacb1"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Điểm trung bình (50 người): 500.40\n",
            "Độ lệch chuẩn (50 người): 8.18\n",
            "Miền giá trị (50 người): [478.60, 517.14]\n",
            "Miền phân vị (IQR) (50 người): 10.32\n",
            "Tỷ lệ đạt trên 450 điểm (50 người): 100.00%\n",
            "\n",
            "Điểm trung bình (100 người): 499.28\n",
            "Độ lệch chuẩn (100 người): 8.20\n",
            "Tỷ lệ đạt trên 450 điểm (100 người): 100.00%\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import scipy.stats as stats\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Tham số của phân phối chuẩn\n",
        "mu = 500  # Kỳ vọng\n",
        "sigma = 8  # Độ lệch chuẩn\n",
        "\n",
        "# a. Mô phỏng ngẫu nhiên điểm thi TOEIC của 50 người\n",
        "group_50 = np.random.normal(mu, sigma, 50)\n",
        "\n",
        "# b. Tính điểm trung bình và độ lệch chuẩn\n",
        "mean_score_50 = np.mean(group_50)\n",
        "std_score_50 = np.std(group_50, ddof=1)\n",
        "\n",
        "# c. Tìm miền giá trị và miền phân vị (IQR)\n",
        "min_value_50 = np.min(group_50)\n",
        "max_value_50 = np.max(group_50)\n",
        "Q1_50 = np.percentile(group_50, 25)\n",
        "Q3_50 = np.percentile(group_50, 75)\n",
        "IQR_50 = Q3_50 - Q1_50\n",
        "\n",
        "# d. Tỷ lệ đạt trên 450 điểm\n",
        "passing_rate_50 = np.mean(group_50 > 450)\n",
        "\n",
        "# e. Vẽ histogram và hàm mật độ\n",
        "plt.figure(figsize=(12, 6))\n",
        "plt.hist(group_50, bins=10, density=True, alpha=0.6, color='g', label='Histogram')\n",
        "\n",
        "# Hàm mật độ chuẩn\n",
        "x = np.linspace(mu - 4*sigma, mu + 4*sigma, 1000)\n",
        "pdf = stats.norm.pdf(x, mu, sigma)\n",
        "plt.plot(x, pdf, 'k', linewidth=2, label='Hàm mật độ chuẩn')\n",
        "\n",
        "plt.title('Histogram điểm thi TOEIC của 50 người')\n",
        "plt.xlabel('Điểm thi')\n",
        "plt.ylabel('Mật độ')\n",
        "plt.legend()\n",
        "plt.grid()\n",
        "plt.show()\n",
        "\n",
        "# In kết quả\n",
        "print(f\"Điểm trung bình (50 người): {mean_score_50:.2f}\")\n",
        "print(f\"Độ lệch chuẩn (50 người): {std_score_50:.2f}\")\n",
        "print(f\"Miền giá trị (50 người): [{min_value_50:.2f}, {max_value_50:.2f}]\")\n",
        "print(f\"Miền phân vị (IQR) (50 người): {IQR_50:.2f}\")\n",
        "print(f\"Tỷ lệ đạt trên 450 điểm (50 người): {passing_rate_50:.2%}\")\n",
        "\n",
        "# f. Mô phỏng nhóm với 100 người\n",
        "group_100 = np.random.normal(mu, sigma, 100)\n",
        "\n",
        "# Tính toán cho nhóm 100 người\n",
        "mean_score_100 = np.mean(group_100)\n",
        "std_score_100 = np.std(group_100, ddof=1)\n",
        "passing_rate_100 = np.mean(group_100 > 450)\n",
        "\n",
        "# In kết quả cho nhóm 100 người\n",
        "print(f\"\\nĐiểm trung bình (100 người): {mean_score_100:.2f}\")\n",
        "print(f\"Độ lệch chuẩn (100 người): {std_score_100:.2f}\")\n",
        "print(f\"Tỷ lệ đạt trên 450 điểm (100 người): {passing_rate_100:.2%}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Mopww78OEGFp"
      },
      "source": [
        "**Bài 10**\n",
        "\n",
        "Viết một chương trình để mô phỏng việc tung đồng xu và đánh giá phương sai của giá trị mô phỏng của biến ngẫu nhiên X (tung được mặt `Head`) đối với các số lần tung khác nhau. Giả sử  đồng xu phải cân bằng.\n",
        "\n",
        "Đối với mỗi N từ 10; 40; 90; 160; 250; 490; 640; 810; 1000, hãy ước tính giá trị của X bằng cách mô phỏng số lần tung đó.\n",
        "\n",
        "Bạn nên chạy mỗi mô phỏng 100 lần và sử dụng tập hợp các ước tính để đánh giá phương sai của ước tính x của bạn. Vẽ biểu đồ phương sai này so với N, bạn có nhận xét gì không?"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 727
        },
        "id": "DQF2fX-_EGFp",
        "outputId": "ee4f340c-3923-4eb6-9d9d-1edbc614660d"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2AAAAIqCAYAAABCJikaAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACEPklEQVR4nO3deXwN9/7H8ffJIhHETqy11KVKpdZq7Xup2qqW2pVSUaStpUXQxVLUEq3qrdLeKqXoYg2l1dLWfquWq2oribURomSb3x/zO4cjCUmcZE6S1/PxmIeTOd8z8zkz55C3mfmMzTAMQwAAAACAdOdhdQEAAAAAkF0QwAAAAAAggxDAAAAAACCDEMAAAAAAIIMQwAAAAAAggxDAAAAAACCDEMAAAAAAIIMQwAAAAAAggxDAACCTiomJkWEYVpcBAABSgQAGAJnQ4MGDlTt3bhUuXFhbt261uhyH69ev64033tDmzZutLgVZ1K+//qpixYpp9erVVpcCAGlCAAOQKosWLZLNZtOuXbusLiXTKFOmjPr06ePSZZYoUUKrVq1S8+bNtXLlSpcu+34MHjxYK1euVO3atV22zPTYfllB69atNWDAAKvLcIkrV67o4sWL2rt3r2w2m0JDQ3Xt2jWnMYcOHZK3t7c6dOigtWvXatWqVTp//rxFFbvWY489ppEjR973cvr06aPcuXO7oCIA6YkABkDSrWBln3x9ffWvf/1LQUFBOnfunNXl4Q5jx45V5cqVde7cOb300ktWlyNJ+uijj/T9999r7dq1ypMnj9XluKWxY8fKZrMledRy6dKljvBxLz/99JM2btyoUaNGOeYdPHhQEyZM0IkTJ1xYccZo166dChcurOrVq0uShg4dqqCgIKcxL7/8sgYNGqQKFSpo5cqVWrx4sYoUKWJFuS43atQozZs3TxEREfcce/36dU2YMMGtjnwnx/7viq+vr86cOZPo+UaNGqlKlSoWVAZYiwAGwMmkSZP06aefKjQ0VI8//rjef/991a1bV9evX7e6tEzryJEj+vDDD12+3NDQUE2aNEkPPvigy5edWvHx8bp06ZLWr1+vYsWKWV2O2xo7dqzKly+vQYMGKSYmxjE/MjJSI0aMUK1atfTiiy/ecznvvPOOmjZt6rTvDx48qIkTJ2bKADZjxgyFhYXpP//5jyTp1VdfdToitGHDBh09elRTp07Vxx9/rPnz5+vkyZNWlety7dq1k7+/v9577717jr1+/bomTpyYKQKY3c2bNzVlyhSrywDcBgEMgJMnn3xSPXr00PPPP69FixZp+PDhOn78uL766iuXr2vHjh2qWbPmPcdVrFhRv//+u8vXn1F8fHzk7e3t8uXOmDFD9erVc/lyU2Ly5Ml65ZVXHD97enpq5MiRqlSpkmNeeHi4ChcubEV5atmypb799tu7jvn888/VuXPnDKrI5Ovrq/fff19HjhzR5MmTHfNHjx6tCxcuaMGCBfLwuPs/zefPn9eaNWv07LPPpne5GaZGjRpq1qyZnnjiCUlS5cqVVblyZcfzLVu21NGjR+Xn56eyZcvqwoULeuCBB6wq1+U8PDz0zDPP6JNPPsmSjXUCAwP14Ycf6uzZs1aXArgFAhiAu2rSpIkk6fjx407zb968qeDgYBUuXFi5cuVShw4ddOHCBacxNptNEyZMSLRMm82mUqVK6cKFC44jaxcvXtSAAQNUtGhR+fr6qkqVKqpXr57Onz+vGzduKDo6WidOnJDNZtP06dO1YMEClS9fXj4+PqpVq5Z27tyZaD3Lly9X5cqVHctbtWqV+vTpozJlytzzfe/atUstW7ZUoUKFlDNnTpUtW1b9+vVzGjN9+nQ9/vjjKliwoHLmzKkaNWpoxYoViZaV0muYEhISNHv2bFWtWlW+vr4qXLiwWrVq5bjezv7+Fy1alOQ2vX1bnzx5Ui+++KIqVqyonDlzqmDBgurcuXOKjo5s3bo1ydPk7Ot//fXXdenSJce+279/v1q3bi1/f3/lzp1bAQEBeuuttxQREaHo6GjH6+1HecqUKSMfHx+VLFlSvXr10sWLF+9Z0+0mTJggm82WaL79dKeff/5Z165dc9T35ZdfqkaNGsqZM6cKFSqkggULavPmzbp48aKjPvu1M2fOnFH79u0dDU5eeeUVxcfHO63n0qVL6tmzp/z9/ZUvXz717t1b+/fvT3bf3Kl58+bq3r27Jk+erP/973/asWOHFixYoGHDhikwMPCer1+zZo3i4uLUrFkzp/duD5ONGzd2nEps34fJfRfv/Gzat+FPP/10z+93QkKCJkyYoOLFi8vPz0+NGzfWwYMHU/x5j4yMVJ8+fZQ3b17HdoyMjExy7Hfffaf69esrV65cypcvn9q1a6dDhw45jbF/Lv744w/16dNH+fLlU968edW3b99ER/D/+ecfvfTSSypUqJDy5Mmjp59+WmfOnEl2O92ud+/e8vX1TbT+li1bKn/+/I6Qca/P6Z3fxebNm+vkyZPat29fsus+ceKE4z81Jk6c6NjPd9acks9xQkKCZs2apYcffli+vr4qWrSoXnjhBf3999+J1vvee+/p4Ycflo+Pj4oXL64hQ4Yku6+S8tprryk+Pp6jYMD/87K6AADu7dixY5KkggULOs0fOnSo8ufPr5CQEJ04cUKzZs1SUFCQli1blqLl/vXXX2rXrp2GDRumGzduqHHjxjpy5IiCgoJUtmxZffHFF/rxxx9VtGhRFS9eXFWqVHFccL9kyRJdvXpVL7zwgmw2m6ZNm6aOHTvqzz//dBxpWrNmjbp06aKqVatq8uTJ+vvvv9W/f3+VKFHinrWdP39eLVq0UOHChTV69Gjly5dPJ06cSNTsYvbs2Xr66af13HPPKSYmRkuXLlXnzp317bffqk2bNinaDrfr37+/Fi1apCeffFLPP/+84uLitG3bNv38888pOlJ4u507d2r79u3q2rWrSpYsqRMnTuj9999Xo0aNdPDgQfn5+aW6Pru3335bNptNX375pQ4dOqT69esrd+7cGjlypLy9vTV27FjH1K5dO0nStWvXVL9+fR06dEj9+vVT9erVdfHiRX399df666+/VKhQoTTXc6e6desqT548qlu3rv7zn/+oZ8+eqlWrliZPnqxz585pypQpjvDy7rvvOl4XHx+vli1bqk6dOpo+fbo2bdqkGTNmqHz58ho8eLAk85fWtm3b6tdff9XgwYNVqVIlffXVV+rdu3eqapw5c6bWrVunF154QZcuXVLJkiU1ceLEFL12+/btKliwoNMRoAYNGuill17SnDlz9Nprr+mhhx6SJMefqZWS7/eYMWM0bdo0tW3bVi1bttT+/fvVsmVL3bhx457LNwxD7dq1048//qhBgwbpoYce0qpVq5Lcjps2bdKTTz6pcuXKacKECfrnn380d+5cPfHEE9qzZ0+i/1B59tlnVbZsWU2ePFl79uzRv//9bxUpUkRTp051jOnTp4+++OIL9ezZU4899pi+//77FH9nZ8+ere+++069e/fWjh075OnpqQ8++EAbN27Up59+quLFi6doOXeqUaOGJPP6vkcffTTJMYULF9b777+vwYMHq0OHDurYsaMk6ZFHHnGMScnnWJJeeOEFLVq0SH379tVLL72k48ePKzQ0VHv37tVPP/3k+Lt0woQJmjhxopo1a6bBgwfryJEjev/997Vz506ncXdTtmxZ9erVSx9++KFGjx6d5m0EZBkGABiG8fHHHxuSjE2bNhkXLlwwTp8+bSxdutQoWLCgkTNnTuOvv/5yGtesWTMjISHB8foRI0YYnp6eRmRkpGOeJCMkJCTRuh544AGjR48exrFjx4yEhARj9uzZhiTjP//5j2NMTEyMUbduXUOScebMGcMwDOP48eOGJKNgwYLG5cuXHWO/+uorQ5LxzTffOOZVrVrVKFmypHH16lXHvK1btxqSjAceeOCu22LVqlWGJGPnzp13HXf9+nWnn2NiYowqVaoYTZo0SfR+e/fufddlfffdd4Yk46WXXkr0nH0729//xx9/nGjMndv6ztoMwzB27NhhSDI++eSTu9ayZcsWQ5KxZcsWp/n29YeGhhqnT582DMMwOnToYOTIkcM4duyYY9zZs2eNXLlyGSVKlDBiY2MNwzCM8ePHG5KMlStXJvv+knPn9gsJCTGS+ufL/tn85ZdfjMuXLxsxMTFG0aJFjSpVqhj//POPY9y3335rSDJ69uzpmNe7d29DkjFp0iSnZT766KNGjRo1HD9/+eWXhiRj1qxZjnnx8fFGkyZNkt03yfnggw8MSYYkY/Xq1Sl+Xb169Zxqslu+fHmS+80w7v5dvH3bpvT7HRERYXh5eRnt27d3Wt6ECRMMSff8vK9evdqQZEybNs0xLy4uzqhfv36i7RgYGGgUKVLEuHTpkmPe/v37DQ8PD6NXr16OefbPRb9+/ZzW1aFDB6NgwYKOn3fv3m1IMoYPH+40rk+fPslupztt2LDBkGS8+eabxp9//mnkzp070ba41+f0+PHjiZ7LkSOHMXjw4Luu+8KFC8nWmdLP8bZt2wxJxmeffeY0bv369U7zz58/b+TIkcNo0aKFER8f7xgXGhpqSDIWLlx411rt73Xnzp3GsWPHDC8vL6e/4xo2bGg8/PDDd10GkBVxCiIAJ82aNVPhwoVVqlQpde3aVblz59aqVasSHTkaOHCg0+k19evXV3x8fIovjPf09FS5cuVks9m0Zs0aBQQEqFu3bo7nvb29Hd399uzZ4/TaLl26KH/+/E7rlqQ///xTknT27Fn99ttv6tWrl1NL5oYNG6pq1ar3rC1fvnySpG+//VaxsbHJjsuZM6fj8d9//60rV66ofv36iepNiS+//FI2m00hISGJnkvqNKZ7ub222NhYXbp0SQ8++KDy5cuXpvpulytXLpUsWVLx8fHasGGD2rdvr3LlyjmeL1asmHr06KHw8HCn0wCrVaumDh06JFpeWt7f3RQpUkT58+fXrl27dO7cOb344ovy9fV1PN+mTRtVqlRJBw8eTPTaQYMGOf1cv359x+dKktavXy9vb2+n9u8eHh4aMmRIquu0H/Xz8/NL1bV8ly5dcvr8p4d7fb83b96suLi4RA1Dhg4dmqLlr127Vl5eXk5HZDw9PRO9Pjw8XPv27VOfPn1UoEABx/xHHnlEzZs319q1axMtO6l9eOnSJUVFRUky96GkNNcuSS1atNALL7ygSZMmqWPHjvL19dUHH3yQ4tcnJ3/+/Kk+JTcp9/ocL1++XHnz5lXz5s118eJFx1SjRg3lzp1bW7ZskWQefYyJidHw4cOdrk0cMGCA/P39tWbNmhTXVK5cOfXs2VMLFixQeHj4fb5DIHMjgAFwMm/ePIWFhWnLli06ePCg/vzzT7Vs2TLRuNKlSzv9bP+FMKnrB+7l5MmTqlChQqLmA/bTp+4Mdfdat318Ut0BU9IxsGHDhurUqZMmTpyoQoUKqV27dvr444918+ZNp3HffvutHnvsMfn6+qpAgQKO04OuXLlyz3Xc6dixYypevLjTL5n3459//tH48eNVqlQp+fj4qFChQipcuLAiIyPTVF9S7NfwVaxYMdFzDz30kBISEnT69GlJ5vvL6HbT9s9BUvVVqlQp0efKft3d7fLnz+/0mT558qSKFSuW6BTO1HaivHr1ql566SVVrFhRMTExTu3kU8JI50YNaf2OFShQIEXh0L4d77xn1Z376m778KGHHnK6ji81tXt4eKhs2bJO41K7D6dPn64CBQpo3759mjNnjkta4huGcd//IZGSz/HRo0d15coVFSlSRIULF3aarl275jjdO7ntnyNHDpUrVy7VnSjHjh2ruLg4rgVDtsc1YACc1K5dO0XXG3l6eiY5PyW/GN55MXhq3c+6U8Jms2nFihX6+eef9c0332jDhg3q16+fZsyYoZ9//lm5c+fWtm3b9PTTT6tBgwZ67733VKxYMXl7e+vjjz/WkiVLXFJHUnUlJantOXToUH388ccaPny46tatq7x588pms6lr165KSEhw2XqskF71Jfe5Sg+vv/66IiIi9Ouvv2rp0qWaPn26+vbt6+gCeDcFCxZM0390JCW5bZbe37H0lFG179271xFUfvvtN6cj+FLaPqeRkZH3fT1kSj7HCQkJKlKkiD777LMkn0+v7qXlypVTjx49tGDBAo0ePTpd1gFkBhwBA5Bu8ufPn6hTVkxMTKLTTx544AEdPXo0UTA4fPiw4/nUsI//448/Ej2X1LzkPPbYY3rrrbe0a9cuffbZZ/r999+1dOlSSeYpdb6+vo5w9uSTTzp1pUut8uXL6+zZs7p8+XKyY+z/k3/nNk3qf6FXrFih3r17a8aMGXrmmWfUvHlz1atXL0Wdy1K6nsKFC8vPz09HjhxJtIzDhw/Lw8NDpUqVkmS+vwMHDtxz3SmR0vrsn4Ok6jty5Eia2pg/8MADTqdW2qXmc7Vr1y7NmzdPQUFBql69ukJCQlSqVCkNGjRIcXFx93x9pUqVEnUlle5+KmdKv4spldx37NKlSykKh/bteO3aNaf5d+6ru+3Dw4cPq1ChQsqVK1eqa09ISEi0DVOzD6Ojo9W3b19VrlxZAwcO1LRp0xJ1Yk3N91UyOxfGxMTcs3GKK07ZLV++vC5duqQnnnhCzZo1SzRVq1ZNUvLbPyYmRsePH0/Td8h+FOz2pihAdkMAA5Buypcvrx9++MFp3oIFCxL9D/BTTz2liIgIpw5rcXFxmjt3rnLnzq2GDRumar32romffPKJ0y9433//vX777bd7vv7vv/9O9L/l9vbg9tMQPT09ZbPZnN7LiRMntHr16lTVatepUycZhpFkJzx7Lf7+/ipUqFCibZrUzVs9PT0TvYe5c+em6CjRAw88IE9Pz3uux9PTU61atdJXX33l1FL73LlzWrJkierVqyd/f3/H+9u/f79WrVqV7PtLqfLly0uSU33R0dFavHix07iaNWuqaNGimj9/vtPpo+vWrdOhQ4fS1KmyZcuWio2NdbqxdkJCgubNm5ei18fHx+uFF15QsWLF9MYbb0gyr6mbO3euDhw44NSVMTl169bV33//7XRNj305UuJf+KWUfxdTqmnTpvLy8tL777/vND80NDRFr2/durXi4uKcXh8fH6+5c+c6jStWrJgCAwO1ePFip/d14MABbdy4Ua1bt0517fZTqu/8PN+57rsZNWqUTp06pcWLF2vmzJkqU6aMevfu7fQ5S+nn1G737t2SpMcff/yu67af/pqaNvB3evbZZxUfH+/4DN4uLi7OsexmzZopR44cmjNnjtP39KOPPtKVK1fS9B0qX768evTooQ8++EARERFpfg9AZsYpiADSzfPPP69BgwapY8eOatGihfbv36/169cnamk/YMAALViwQH369NHu3btVpkwZrVixQj/99JNmzZqlPHnypHrdb7/9ttq1a6cnnnhCffv21d9//63Q0FBVqVIl0f+632nx4sV677331KFDB5UvX15Xr17Vhx9+KH9/f8cvfG3atNHMmTPVqlUrde/eXefPn9e8efP04IMP6r///W+q623cuLF69uypOXPm6OjRo2rVqpUSEhK0bds2NW7cWEFBQZLMbTplyhQ9//zzqlmzpn744Qf973//S7S8p556Sp9++qny5s2rypUra8eOHdq0aVOibZ+UvHnzqnPnzpo7d65sNpvKly+vb7/9Nslflt58801t3LhR9erV04svvigvLy998MEHunnzpqZNm+YY9+qrr2rFihXq3Lmz+vXrpxo1aujy5cv6+uuvNX/+fMf/uKdEixYtVLp0afXv31+vvvqqPD09tXDhQhUsWFCnTp1yjPP29tY777yjXr16qWHDhurWrZvOnTun2bNnq0yZMhoxYkSK12nXvn171a5dWy+//LL++OMPVapUSV9//bXjyOW9jk7MmTNHe/bs0Zdffun0uX766af19NNPa+LEierSpUui65hu16ZNG3l5eWnTpk0aOHCgY35gYKA8PT01depUXblyRT4+PmrSpImKFCni+C526tRJzZs31/79+7Vhw4Y0n+5WtGhRDRs2TDNmzNDTTz+tVq1aaf/+/Vq3bp0KFSp0z+3Qtm1bPfHEExo9erROnDihypUra+XKlUlen/jOO+/oySefVN26ddW/f39HG/q8efPe855dSalRo4Y6deqkWbNm6dKlS4429Pbv0b1q/+677/Tee+8pJCRE1atXlyR9/PHHatSokcaNG+f43Cf3OS1cuLDT59QuLCxMpUuXTrYFvV3OnDlVuXJlLVu2TP/6179UoEABValSJVXXWDZs2FAvvPCCJk+erH379qlFixby9vbW0aNHtXz5cs2ePVvPPPOMChcurDFjxmjixIlq1aqVnn76aR05ckTvvfeeatWqpR49eqR4nbd7/fXX9emnn+rIkSN6+OGH07QMIFOzpvkiAHdze7vgtIxLqnV5fHy8MWrUKKNQoUKGn5+f0bJlS+OPP/5Isi37hQsXjP79+xuFCxc2cuTIYVStWjVRS297G/R33nknUV1Koi3z0qVLjUqVKhk+Pj5GlSpVjK+//tro1KmTUalSpbu+xz179hjdunUzSpcubfj4+BhFihQxnnrqKWPXrl1O4z766COjQoUKho+Pj1GpUiXj448/TrL1dEra0BuG2Yb7nXfeMSpVqmTkyJHDKFy4sPHkk08au3fvdoy5fv260b9/fyNv3rxGnjx5jGeffdY4f/58ovf/999/G3379jUKFSpk5M6d22jZsqVx+PDhFNdy4cIFo1OnToafn5+RP39+44UXXjAOHDiQZKv1vXv3Gq1atTJy585t+Pn5GY0bNza2b9+eaJmXLl0ygoKCjBIlShg5cuQwSpYsafTu3du4ePHiXWtJqubdu3cbderUMXLkyGGULl3amDlzZrLtvZcvX25Ur17d8PHxMQoUKGA899xzjtsq2PXu3dvIlStXonUntT8vXLhgdO/e3ciTJ4+RN29eo0+fPsZPP/1kSDKWLl2a7Ps4ffq0kTt3buOpp55K8vmTJ08auXLlMp5++um7bA3T008/bTRt2jTR/A8//NAoV66c4enp6fR9TOl3MTXf77i4OGPcuHFGQECAkTNnTqNJkybGoUOHjIIFCxqDBg2653u4dOmS0bNnT8Pf39/Imzev0bNnT2Pv3r1JfsY2bdpkPPHEE0bOnDkNf39/o23btsbBgwedxtj31YULF5zmJ/W5iI6ONoYMGWIUKFDA0UL+yJEjhiRjypQpydYcFRVlPPDAA0b16tUdt1iwGzFihOHh4WHs2LHDMS+ln9P4+HijWLFixtixY++53QzDMLZv327UqFHDyJEjh9N3PzWfY8MwjAULFhg1atQwcubMaeTJk8eoWrWqMXLkSOPs2bNO40JDQ41KlSoZ3t7eRtGiRY3Bgwcbf//99z3rvNu/K/aW+bShR3ZkM4xMcEUtALhIYGCgChcurLCwMKtLQRayevVqdejQQT/++GOKGmncr23btqlRo0Y6fPiwKlSokO7rS6nIyEjlz59fb775pl5//XWry0mVffv26dFHH9V//vMfPffccxm67tWrV6t79+46duyYihUrlqHrBpDxuAYMQJYUGxubqKHB1q1btX//fjVq1MiaomApm82WplPW7vTPP/84/Wy/dsnf399xSlp6q1+/vlq0aOF0mmdGu3M7SNKsWbMkye2/Y8nV7uHhoQYNGmR4PVOnTlVQUBDhC8gmuAYMQJZ05swZNWvWTD169FDx4sV1+PBhzZ8/XwEBAYluUgqkxtChQ/XPP/+obt26unnzplauXKnt27fr7bffdroBdnpbt25dhq0rKcuWLdOiRYvUunVr5c6dWz/++KM+//xztWjRIkOOAt6PadOmaffu3WrcuLG8vLy0bt06rVu3TgMHDnR07sxIO3bsyPB1ArAOpyACyJKuXLmigQMH6qefftKFCxeUK1cuNW3aVFOmTHF0J0P2cuPGDXl5ecnL6/7+73HJkiWaMWOG/vjjD924cUMPPvigBg8e7GiUkl3s2bNHI0eO1L59+xQVFaWiRYuqU6dOevPNNxPdYNndhIWFaeLEiTp48KCuXbum0qVLq2fPnnr99dfv+/MBAPdCAAMAAACADMI1YAAAAACQQQhgAAAAAJBBONH5PiQkJOjs2bPKkyfPPW/cCAAAACDrMgxDV69eVfHixeXhkfxxLgLYfTh79qwl3ZIAAAAAuKfTp0+rZMmSyT5PALsPefLkkWRuZH9/f0triY2N1caNG9WiRQt5e3tbWguQ2fF9yl7Y3+mL7euM7eG+MsO+cbca3aEed6jBLioqSqVKlXJkhOQQwO6D/bRDf39/twhgfn5+8vf3t/zDB2R2fJ+yF/Z3+mL7OmN7uK/MsG/crUZ3qMcdarjTvS5NogkHAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYAAAAAGQQtwlg8+bNU5kyZeTr66s6dero119/vev45cuXq1KlSvL19VXVqlW1du1ax3OxsbEaNWqUqlatqly5cql48eLq1auXzp4967SMMmXKyGazOU1TpkxJl/eXnuLjpe+/t+mHH0ro++9tio+3uiIAAAAASXGLALZs2TIFBwcrJCREe/bsUbVq1dSyZUudP38+yfHbt29Xt27d1L9/f+3du1ft27dX+/btdeDAAUnS9evXtWfPHo0bN0579uzRypUrdeTIET399NOJljVp0iSFh4c7pqFDh6bre3W1lSulMmWk5s29NHNmTTVv7qUyZcz5AAAAANyLl9UFSNLMmTM1YMAA9e3bV5I0f/58rVmzRgsXLtTo0aMTjZ89e7ZatWqlV199VZL0xhtvKCwsTKGhoZo/f77y5s2rsLAwp9eEhoaqdu3aOnXqlEqXLu2YnydPHgUEBKSozps3b+rmzZuOn6OioiSZR9xiY2NT96ZdYNUqm7p29ZRhSJLNMf/MGUPPPCMtXRqvDh2MDK8LyOzs32crvtfIeOzv9MX2dcb2cF+ZYd+4W43uUI871GCX0hpshmFY+ht6TEyM/Pz8tGLFCrVv394xv3fv3oqMjNRXX32V6DWlS5dWcHCwhg8f7pgXEhKi1atXa//+/UmuZ9OmTWrRooUiIyPl7+8vyTwF8caNG4qNjVXp0qXVvXt3jRgxQl5eSefSCRMmaOLEiYnmL1myRH5+fql41/cvPl4aOLCFLl3y1e3h6xZDhQr9ow8+CJOnZ4aWBgAAAGQ7169fV/fu3XXlyhVH3kiK5UfALl68qPj4eBUtWtRpftGiRXX48OEkXxMREZHk+IiIiCTH37hxQ6NGjVK3bt2cNsZLL72k6tWrq0CBAtq+fbvGjBmj8PBwzZw5M8nljBkzRsHBwY6fo6KiVKpUKbVo0eKuGzk9fP+9TZcu3W332XTxop/8/duoYUOOggGpERsbq7CwMDVv3lze3t5Wl4N0xv5OX2xfZ2wP95UZ9o271egO9bhDDXb2s+PuxfIAlt5iY2P17LPPyjAMvf/++07P3R6mHnnkEeXIkUMvvPCCJk+eLB8fn0TL8vHxSXK+t7d3hu/wCxdSOs5LbvD9BDIlK77bsA77O32xfZ2xPdxXZtg37lajO9TjLjWkhOVNOAoVKiRPT0+dO3fOaf65c+eSvTYrICAgRePt4evkyZMKCwu751GqOnXqKC4uTidOnEj9G8lgxYq5dhwAAACA9Gd5AMuRI4dq1KihzZs3O+YlJCRo8+bNqlu3bpKvqVu3rtN4SQoLC3Mabw9fR48e1aZNm1SwYMF71rJv3z55eHioSJEiaXw3Gad+falkScmW1OVfMueXKmWOAwAAAOAe3OIUxODgYPXu3Vs1a9ZU7dq1NWvWLEVHRzu6Ivbq1UslSpTQ5MmTJUnDhg1Tw4YNNWPGDLVp00ZLly7Vrl27tGDBAklm+HrmmWe0Z88effvtt4qPj3dcH1agQAHlyJFDO3bs0C+//KLGjRsrT5482rFjh0aMGKEePXoof/781myIVPD0lGbPlp55xgxbd7ZSMQxp1izRgAMAAABwI24RwLp06aILFy5o/PjxioiIUGBgoNavX+9otHHq1Cl5eNw6WPf4449ryZIlGjt2rF577TVVqFBBq1evVpUqVSRJZ86c0ddffy1JCgwMdFrXli1b1KhRI/n4+Gjp0qWaMGGCbt68qbJly2rEiBFO14W5u44dpRUrpGHDpL/+cn7Oy0u6460DAAAAsJhbBDBJCgoKUlBQUJLPbd26NdG8zp07q3PnzkmOL1OmjO7VXb969er6+eefU12nu+nYUWrXTtqyJU7r1u1Tq1aBmjzZS1u2SMOHS/+fQwEAAAC4AcuvAcP98/SUGjY01KDBGTVqZGjePPMI2DffSN9+a3V1AAAAAOwIYFnQQw+ZR78k8/TEGzcsLQcAAADA/yOAZVHjx0vFi0t//ilNm2Z1NQAAAAAkAliWlSePNGOG+XjyZOn4cWvrAQAAAEAAy9K6dJEaNzZPQbSfkggAAADAOgSwLMxmk+bONRtyfP21tGaN1RUBAAAA2RsBLIt7+GGzEYckvfQSDTkAAAAAKxHAsoGQkFsNOd55x+pqAAAAgOyLAJYN5MkjTZ9uPn77benECUvLAQAAALItAlg20bWr1LAhDTkAAAAAKxHAsgmbTZo3T/L0lL76Slq71uqKAAAAgOyHAJaN0JADAAAAsBYBLJsJCZGKFZOOHbt1XRgAAACAjEEAy2b8/WnIAQAAAFiFAJYNdetmNuT45x9pxAirqwEAAACyDwJYNmSzSaGhZkOO1auldeusrggAAADIHghg2VSVKmYjDsn88+ZNa+sBAAAAsgMCWDY2YYIUECD98QcNOQAAAICMQADLxm5vyPHWW9LJk9bWAwAAAGR1BLBsrnt3qUEDGnIAAAAAGYEAls3d3pBj1Spp/XqrKwIAAACyLgIYVLWqNHSo+XjoUBpyAAAAAOmFAAZJzg05ZsywuhoAAAAgayKAQZKUN6/0zjvm4zfflE6dsrYeAAAAICsigMHhueek+vVpyAEAAACkFwIYHG5vyLFypbRhg9UVAQAAAFkLAQxOHnlECgoyH9OQAwAAAHAtAhgSmThRKlpUOnpUmjnT6moAAACArIMAhkRoyAEAAACkDwIYktSjh1SvnnT9uhQcbHU1AAAAQNZAAEOSbDZp3jyzIceXX0obN1pdEQAAAJD5EcCQrEcekYYMMR/TkAMAAAC4fwQw3JW9Icf//ie9+67V1QAAAACZGwEMd5UvnzRtmvn4jTek06ctLQcAAADI1AhguKeePaUnnqAhBwAAAHC/CGC4J3tDDg8PacUKKSzM6ooAAACAzIkAhhSpVs25IUdMjLX1AAAAAJkRAQwpNmmSVKSIdOQIDTkAAACAtCCAIcVoyAEAAADcHwIYUqVnT+nxx6XoaOnll62uBgAAAMhcCGBIFQ+PWw05li+XNm2yuiIAAAAg8yCAIdUCA6UXXzQf05ADAAAASDkCGNLkjTekwoWlw4elWbOsrgYAAADIHAhgSJPbG3JMmiT99Zel5QAAAACZAgEMadarl1S3Lg05AAAAgJQigCHNbm/I8cUX0ubNVlcEAAAAuDcCGO7Lo49Kgwebj2nIAQAAANwdAQz3zd6Q49AhafZsq6sBAAAA3BcBDPctf35p6lTz8cSJNOQAAAAAkkMAg0v07i099pjZkOOVV6yuBgAAAHBPBDC4xO0NOZYtk7ZssboiAAAAwP0QwOAy1atLgwaZj4cMkWJjra0HAAAAcDcEMLjUm29KhQrRkAMAAABICgEMLnVnQ44zZ6ytBwAAAHAnBDC4XJ8+Up060rVrNOQAAAAAbkcAg8t5eEjvvSfZbNLSpTTkAAAAAOwIYEgXtzfkCAqiIQcAAAAgEcCQjt58UypYUDp4UJozx+pqAAAAAOsRwJBuChS41ZBjwgTp7FlLywEAAAAsRwBDuurbl4YcAAAAgB0BDOnKw0OaN89syPH559LWrVZXBAAAAFiHAIZ0V6OG9MIL5mMacgAAACA7I4AhQ7z1ltmQ4/ffpblzra4GAAAAsAYBDBmiQAFpyhTzMQ05AAAAkF0RwJBh+vWTateWrl6VXn3V6moAAACAjEcAQ4a5vSHHkiXS999bXREAAACQsQhgyFA1a0oDB5qPacgBAACA7IYAhgz31lvmNWEHDkihoVZXAwAAAGQcAhgyXMGCtxpyhIRI4eHW1gMAAABkFAIYLNG/v1SrFg05AAAAkL0QwGCJ2xtyfPaZ9MMPVlcEAAAApD8CGCxTq5Y0YID5eMgQGnIAAAAg6yOAwVJvv32rIce8eVZXAwAAAKQvAhgsVbCgNHmy+TgkRIqIsLYeAAAAID25TQCbN2+eypQpI19fX9WpU0e//vrrXccvX75clSpVkq+vr6pWraq1a9c6nouNjdWoUaNUtWpV5cqVS8WLF1evXr109uxZp2VcvnxZzz33nPz9/ZUvXz71799f165dS5f3h+T172/eHywqSho50upqAAAAgPTjFgFs2bJlCg4OVkhIiPbs2aNq1aqpZcuWOn/+fJLjt2/frm7duql///7au3ev2rdvr/bt2+vAgQOSpOvXr2vPnj0aN26c9uzZo5UrV+rIkSN6+umnnZbz3HPP6ffff1dYWJi+/fZb/fDDDxpov0swMoynp/Tee2ZDjk8/lbZts7oiAAAAIH14WV2AJM2cOVMDBgxQ3759JUnz58/XmjVrtHDhQo0ePTrR+NmzZ6tVq1Z69f/7l7/xxhsKCwtTaGio5s+fr7x58yosLMzpNaGhoapdu7ZOnTql0qVL69ChQ1q/fr127typmjVrSpLmzp2r1q1ba/r06SpevHii9d68eVM3b950/BwVFSXJPOIWa3EHCfv6ra4jrQIDpX79PPXRRx568UVDv/4aJy+3+HQiO8rs3yekDvs7fbF9nbE93Fdm2DfuVqM71OMONdiltAabYRhGOtdyVzExMfLz89OKFSvUvn17x/zevXsrMjJSX331VaLXlC5dWsHBwRo+fLhjXkhIiFavXq39+/cnuZ5NmzapRYsWioyMlL+/vxYuXKiXX35Zf//9t2NMXFycfH19tXz5cnXo0CHRMiZMmKCJEycmmr9kyRL5+fml4l0jKVFROTRkSFNdvZpD/fv/prZt/7S6JAAAACBFrl+/ru7du+vKlSvy9/dPdpzlxxguXryo+Ph4FS1a1Gl+0aJFdfjw4SRfExERkeT4iGQ6ONy4cUOjRo1St27dHBsjIiJCRYoUcRrn5eWlAgUKJLucMWPGKDg42PFzVFSUSpUqpRYtWtx1I2eE2NhYhYWFqXnz5vL29ra0lvtx9aqHhgyRli+vopCQSgoIsLoiZEdZ5fuElGF/py+2rzO2h/vKDPvG3Wp0h3rcoQY7+9lx92J5AEtvsbGxevbZZ2UYht5///37WpaPj498fHwSzff29rZ8h9u5Uy1p8cIL0sKF0u7dNo0d663Fi62uCNlZZv8+IXXY3+mL7euM7eG+MsO+cbca3aEed6khJSxvwlGoUCF5enrq3LlzTvPPnTungGQOfwQEBKRovD18nTx5UmFhYU5HqQICAhI1+YiLi9Ply5eTXS/S3+0NOT75RPrxR6srAgAAAFzH8gCWI0cO1ahRQ5s3b3bMS0hI0ObNm1W3bt0kX1O3bl2n8ZIUFhbmNN4evo4ePapNmzapYMGCiZYRGRmp3bt3O+Z99913SkhIUJ06dVzx1pBGtWubreklacgQKS7O2noAAAAAV7E8gElScHCwPvzwQy1evFiHDh3S4MGDFR0d7eiK2KtXL40ZM8YxftiwYVq/fr1mzJihw4cPa8KECdq1a5eCgoIkmeHrmWee0a5du/TZZ58pPj5eERERioiIUExMjCTpoYceUqtWrTRgwAD9+uuv+umnnxQUFKSuXbsm2QERGWvyZCl/fum//zWPiAEAAABZgVsEsC5dumj69OkaP368AgMDtW/fPq1fv97RaOPUqVMKDw93jH/88ce1ZMkSLViwQNWqVdOKFSu0evVqValSRZJ05swZff311/rrr78UGBioYsWKOabt27c7lvPZZ5+pUqVKatq0qVq3bq169eppwYIFGfvmkaRChaS33zYfjxsn3XHGKQAAAJApuU0TjqCgIMcRrDtt3bo10bzOnTurc+fOSY4vU6aMUtJdv0CBAlqyZEmq6kTGGTBA+vBDac8eadQoadEiqysCAAAA7o9bHAEDkmJvyCFJixdLP/1kbT0AAADA/SKAwa3VqUNDDgAAAGQdBDC4vcmTpXz5pP37pfu8lRsAAABgKQIY3F7hwjTkAAAAQNZAAEOmMHCgVL26dOWKNHq01dUAAAAAaUMAQ6bg6SnNm2c+XrRIuu1uAgAAAECmQQBDpvHYY1K/fuZjGnIAAAAgMyKAIVOZMsVsyLFvnzR/vtXVAAAAAKlDAEOmUriw9NZb5uOxY6Xz562tBwAAAEgNAhgynRdekB59lIYcAAAAyHwIYMh0bm/I8fHH0o4d1tYDAAAApBQBDJlS3bpS377m4xdflOLjra0HAAAASAkCGDItGnIAAAAgsyGAIdMqUkR6803z8dix0oUL1tYDAAAA3AsBDJnaoEFSYKAUGUlDDgAAALg/AhgytdsbcixcSEMOAAAAuDcCGDK9xx+X+vQxHw8ZQkMOAAAAuC8CGLKEqVOlvHmlvXulDz6wuhoAAAAgaQQwZAm3N+R4/XUacgAAAMA9EcCQZQwaJFWrZjbkGDPG6moAAACAxAhgyDK8vG415PjoI+nnn62tBwAAALgTAQxZyhNPSL17m49pyAEAAAB3QwBDlmNvyLFnj7RggdXVAAAAALcQwJDlFC0qvfGG+fj116WLF62tBwAAALAjgCFLGjxYeuQR6e+/acgBAAAA90EAQ5Z0e0OOf/9b+uUXa+sBAAAAJAIYsrB69aRevczHNOQAAACAOyCAIUubNk3y95d27zaPhAEAAABWIoAhS7u9IceYMTTkAAAAgLUIYMjyXnzxVkOO116zuhoAAABkZwQwZHleXlJoqPn43/+Wfv3V2noAAACQfRHAkC3Ury/17CkZBg05AAAAYB2vtLzo0Ucflc1mSzTfZrPJ19dXDz74oPr06aPGjRvfd4GAq0ybJn31lbRrl/TRR9LAgVZXBAAAgOwmTUfAWrVqpT///FO5cuVS48aN1bhxY+XOnVvHjh1TrVq1FB4ermbNmumrr75ydb1AmgUESJMmmY/HjJEuXbK2HgAAAGQ/aQpgFy9e1Msvv6xt27ZpxowZmjFjhn744Qe98sorio6O1saNGzV27Fi9YW8/B7iJIUOkqlWly5dpyAEAAICMl6YA9sUXX6hbt26J5nft2lVffPGFJKlbt246cuTI/VUHuNjtDTk+/FDaudPaegAAAJC9pCmA+fr6avv27Ynmb9++Xb6+vpKkhIQEx2PAnTRoIPXocashR0KC1RUBAAAgu0hTE46hQ4dq0KBB2r17t2rVqiVJ2rlzp/7973/rtf8/r2vDhg0KDAx0WaGAK9kbcuzcaTbkGDDA6ooAAACQHaQpgI0dO1Zly5ZVaGioPv30U0lSxYoV9eGHH6p79+6SpEGDBmnw4MGuqxRwoWLFzIYcI0ZIo0dLHTtKBQtaXRUAAACyujQFMEl67rnn9NxzzyX7fM6cOdO6aCBDBAWZR78OHJBef12aP9/qigAAAJDV3deNmGNiYvTXX3/p1KlTThOQGXh5SfPmmY8XLDDvDwYAAACkpzQFsKNHj6p+/frKmTOnHnjgAZUtW1Zly5ZVmTJlVLZsWVfXCKSbBg2k556jIQcAAAAyRppOQezTp4+8vLz07bffqlixYrLZbK6uC8gw77wjff219Ouv0sKF0vPPW10RAAAAsqo0BbB9+/Zp9+7dqlSpkqvrATJcsWLSxIlScPCthhwFClhdFQAAALKiNJ2CWLlyZV28eNHVtQCWCQqSHn5YunTJbMgBAAAApIc0BbCpU6dq5MiR2rp1qy5duqSoqCinCchsvL1vNeT44AMacgAAACB9pOkUxGbNmkmSmjZt6jTfMAzZbDbFx8fff2VABmvYUOreXVqyxDwitn275HFffUIBAAAAZ2kKYFu2bHF1HYBbsDfk+OUX6eOPpf79ra4IAAAAWUmqA9iaNWt048YNNWjQQIULF06PmgDLFC9uNuR4+WVp1CipQwcacgAAAMB1Uh3A1q5dq2vXrmnAgAF6/fXXdejQIUnSww8/rH79+ilv3rwuLxLISEOHmu3of/9dGjtWeu89qysCAABAVpHqK1zmzZunJ598UpGRkXr33Xd1+fJlXb58WTNnzlT58uW1Z8+e9KgTyDDe3lJoqPl4/nxp925r6wEAAEDWkaYWAz169FCLFi104sQJrVy5UitXrtTx48f11FNPafjw4S4uEch4jRpJ3bpJhiENGSIlJFhdEQAAALKCNAUwb29vzZo1S15et85g9PLy0siRI7Xr//t3t2nTRuHh4a6pErDA9OlS7txmQ45Fi6yuBgAAAFlBmgKYv7+/Tp06lWj+6dOnlSdPHklS1apV5ePjc3/VARYqXlwKCTEfjxolXb5sbT0AAADI/NIUwLp06aL+/ftr2bJlOn36tE6fPq2lS5fq+eefV7du3SRJU6ZMUQHaxyGTGzZMqlxZunhRGjfO6moAAACQ2aXpPmDTp0+XzWZTr169FBcXJ8k8LXHw4MGaMmWKSwsErGRvyNGkifT+++Z9wapXt7oqAAAAZFapDmDDhw9XXFycOnXqpMmTJ+uPP/6QzWZT+fLl5efnlx41ApZq3Fjq2lVautRsyPHTT5JHmo4dAwAAILtL9a+R9evXV/Xq1dWlSxf5+fnpkUceUdWqVQlfyNLsDTl+/llavNjqagAAAJBZpfoIWKdOneTh4SGbzaYmTZokOea7776778IAd1KihDR+vDRypDm1by/lz291VQAAAMhs0nQiVfv27fX888+rWrVqqlatmipXrqyYmBjt3r1bVatWdXWNgFsYNkx66CEacgAAACDt0tSEY+XKlUnOnzBhgq5du3ZfBQHuKkcOsyFH06a3GnI8+qjVVQEAACAzcWkrgR49emjhwoWuXCTgVpo0kbp0kRISzIYcCQlWVwQAAIDMxKUBbMeOHfL19XXlIgG3M326lCuXtGOH9MknVlcDAACAzCRNpyB27NjR6WfDMBQeHq5du3ZpHBfHIIsrWdJsyDFqlNmQo107GnIAAAAgZdJ0BCxv3rxOU4ECBdSoUSOtXbtWISEhrq4RcDvDh0uVKkkXLphhDAAAAEiJNB0B+/jjj11dB5Cp2BtyNGsmvfee2ZAjMNDqqgAAAODuXHoNGJCdNG0qPfssDTkAAACQcik+Avboo4/KZrOlaOyePXvSXBCQmcyYIa1ZI23fLn36qdS7t9UVAQAAwJ2lOIC1b98+HcsAMqeSJc2bMo8eLb36qtmQI18+q6sCAACAu0pxAKO5BpC0ESOkjz+WjhwxG3LMmWN1RQAAAHBXXAMG3Cd7Qw5JmjdP2r/f2noAAADgvghggAs0ayZ17nyrIYdhWF0RAAAA3BEBDHCRGTMkPz/pp5/MhhwAAADAnQhggIuUKmU25JDMhhyRkZaWAwAAADfkFgFs3rx5KlOmjHx9fVWnTh39+uuvdx2/fPlyVapUSb6+vqpatarWrl3r9PzKlSvVokULFSxYUDabTfv27Uu0jEaNGslmszlNgwYNcuXbQjYUHCxVrCidPy/RtwYAAAB3SlMA69Spk6ZOnZpo/rRp09S5c+dULWvZsmUKDg5WSEiI9uzZo2rVqqlly5Y6f/58kuO3b9+ubt26qX///tq7d6/at2+v9u3b68CBA44x0dHRqlevXpI13m7AgAEKDw93TNOmTUtV7cCdcuSQ5s41H4eG0pADAAAAzlLchv52P/zwgyZMmJBo/pNPPqkZM2akalkzZ87UgAED1LdvX0nS/PnztWbNGi1cuFCjR49ONH727Nlq1aqVXn31VUnSG2+8obCwMIWGhmr+/PmSpJ49e0qSTpw4cdd1+/n5KSAgIMW13rx5Uzdv3nT8HBUVJUmKjY1VbGxsipeTHuzrt7oOSI0aSR07emrlSg8NGZKg776LVwrvYQ43wfcpe2F/py+2rzO2h/vKDPvG3Wp0h3rcoQa7lNZgM4zU92vLmTOn9u3bp4oVKzrNP3z4sB599FH9888/KVpOTEyM/Pz8tGLFCqcbPffu3VuRkZH66quvEr2mdOnSCg4O1vDhwx3zQkJCtHr1au2/43DDiRMnVLZsWe3du1eBgYFOzzVq1Ei///67DMNQQECA2rZtq3HjxsnPzy/ZeidMmKCJEycmmr9kyZK7vg7Zz4ULvgoKaqqbN700bNhuNW78l9UlAQAAIB1dv35d3bt315UrV+Tv75/suDQdAatataqWLVum8ePHO81funSpKleunOLlXLx4UfHx8SpatKjT/KJFi+rw4cNJviYiIiLJ8RERESleryR1795dDzzwgIoXL67//ve/GjVqlI4cOaKVK1cm+5oxY8YoODjY8XNUVJRKlSqlFi1a3HUjZ4TY2FiFhYWpefPm8vb2trQWmCIibBo3Tlq6tLrGjn1EefNaXRFSiu9T9sL+Tl9sX2dsD/eVGfaNu9XoDvW4Qw129rPj7iVNAWzcuHHq2LGjjh07piZNmkiSNm/erM8//1zLly9PyyIz3MCBAx2Pq1atqmLFiqlp06Y6duyYypcvn+RrfHx85OPjk2i+t7e35Tvczp1qye5efdVsR/+//9n05pvemjXL6oqQWnyfshf2d/pi+zpje7ivzLBv3K1Gd6jHXWpIiTQ14Wjbtq1Wr16tP/74Qy+++KJefvll/fXXX9q0aZPTqYT3UqhQIXl6eurcuXNO88+dO5fstVkBAQGpGp9SderUkST98ccf97UcwM7Hx7khx3//a209AAAAsF6qA9i1a9d06dIltWnTRj/99JOio6N18eJFfffdd2rYsGGqlpUjRw7VqFFDmzdvdsxLSEjQ5s2bVbdu3SRfU7duXafxkhQWFpbs+JSyt6ovVqzYfS0HuF2LFlKnTlJ8vDRkiJT6Ky4BAACQlaT6FMQ6deooLi5OFSpU0KRJk3Tw4EFJ0sMPP6xHH3001QUEBwerd+/eqlmzpmrXrq1Zs2YpOjra0RWxV69eKlGihCZPnixJGjZsmBo2bKgZM2aoTZs2Wrp0qXbt2qUFCxY4lnn58mWdOnVKZ8+elSQdOXJEknn0LCAgQMeOHdOSJUvUunVrFSxYUP/97381YsQINWjQQI888kiq3wNwNzNnSuvWST/+KH32mdSjh9UVAQAAwCopCmCxsbGOcxp///13/fbbb6pWrZrWrVunvP/fWSAyMlKNGzfW0qVLVbhw4RQX0KVLF124cEHjx49XRESEAgMDtX79ekejjVOnTsnD49aBuscff1xLlizR2LFj9dprr6lChQpavXq1qlSp4hjz9ddfOwKcJHXt2lWS2S1xwoQJypEjhzZt2uQIe6VKlVKnTp00duzYFNcNpFTp0tLrr5vTK69IbduKhhwAAADZVIoC2MyZM1WuXDnHTZYfe+wx2Ww2HThwQA899JAk6eDBg+rdu7deeuklff7556kqIigoSEFBQUk+t3Xr1kTzOnfufNcbPvfp00d9+vRJ9vlSpUrp+++/T1WNwP14+WVp0SLp6FFpwgTp3XetrggAAABWSNE1YO3bt9ekSZP07v//1ujl5aWff/7ZEb4kqXLlypo3b57WrVuXPpUCmdjtDTnmzpV++83aegAAAGCNFAWwihUr6ueff9bevXslmY0ycuTIkWict7e3EhISXFshkEW0bCl17EhDDgAAgOwsxV0Qc+XKpU8++USS1KRJEw0bNszR5EKSzpw5oxEjRqhp06aurxLIIt59V8qZU9q2TVqyxOpqAAAAkNHSdB+w0NBQXblyRWXKlFH58uVVvnx5lS1bVlFRUZprP88KQCL2hhyS2ZAjhTdMBwAAQBaR6jb0X331lWJjY7Vx40bt27dPhw8fliQ99NBDatasmcsLBLKaV14xG3L88YfZkGPmTKsrAgAAQEZJdQDbtGmToqKiNGbMGB09elTNmzdPj7qALMvekOPJJ6U5c6R+/aTb7qIAAACALCzVAWzu3Lnq1KmT/vzzT02aNCnJMePHj7/vwoCsrFUrqUMHadUqsyHH1q2SzWZ1VQAAAEhvqQ5gkrRq1Sr5+flp1apVkswbNR8/flyenp568MEHCWBACrz7rrR+vfTDD9Lnn0vdu1tdEQAAANJbmgJYUq3mo6Ki1KdPH7Vv3/5+awKyhQcekF57TRo3zrxR81NPSf7+VlcFAACA9JSmLohJ8ff318SJEzn6BaTCK69IDz4oRURIEydaXQ0AAADSm8sCmCRduXJFV65cceUigSzN19dsxCFJs2dLBw5YWw8AAADSV5pOQZxj/43x/xmGofDwcH366ad68sknXVIYkF08+aTUrp301VdSUJC0ZQsNOQAAALKqNAWwd9991+lnDw8PFS5cWL1799aYMWNcUhiQncyaJW3YIH3/vbR0qdStm9UVAQAAID2kKYAdP37c1XUA2VqZMmZDjvHjzYYcbdrQkAMAACArcuk1YADS7tVXpfLlpfBwKZlb7AEAACCTI4ABbuLOhhy//25tPQAAAHA9AhjgRlq3lp5+WoqLMxtyGIbVFQEAAMCVCGCAm5k1yzwatnWrtGyZ1dUAAADAlQhggJspW1ayNxN9+WXp6lVr6wEAAIDrpKkLot3169d16tQpxcTEOM1/5JFH7qsoILsbOVJavFj680+zIcc771hdEQAAAFwhTQHswoUL6tu3r9atW5fk8/Hx8fdVFJDd2RtyPPWUeUpi375S5cpWVwUAAID7laZTEIcPH67IyEj98ssvypkzp9avX6/FixerQoUK+vrrr11dI5AttWkjtW1LQw4AAICsJE0B7LvvvtPMmTNVs2ZNeXh46IEHHlCPHj00bdo0TZ482dU1AtnW7Nnm0bAtW6QvvrC6GgAAANyvNAWw6OhoFSlSRJKUP39+XbhwQZJUtWpV7dmzx3XVAdlc2bLS6NHm4+BgGnIAAABkdmkKYBUrVtSRI0ckSdWqVdMHH3ygM2fOaP78+SpWrJhLCwSyu5EjpXLlpLNnpTfesLoaAAAA3I80BbBhw4YpPDxckhQSEqJ169apdOnSmjNnjt5++22XFghkdzlzmqciStK770qHDllbDwAAANIuTV0Qe/To4Xhco0YNnTx5UocPH1bp0qVVqFAhlxUHwPTUU+b07bdmQ45NmySbzeqqAAAAkFouuRGzn5+fqlevTvgC0tHs2ZKPj/Tdd9Ly5VZXAwAAgLRI0xGwfv363fX5hQsXpqkYAMkrV85syDFxotmQo3VrKXduq6sCAABAaqTpCNjff//tNJ0/f17fffedVq5cqcjISBeXCMBu1CizM+KZMzTkAAAAyIxSdARs0KBB2rNnjz766CNVrVpVq1atSjQmISFBgwcPVvny5V1eJACTvSHH009LM2dKffpIDz1kdVUAAABIqRQdAfv666+1cOFCzZo1S1eTuRGRh4eHgoOD9e6777q0QADO2raV2rSR4uKkoUMlw7C6IgAAAKRUigJY9+7dFR4ero8++kh58uRJdtyxY8cUFxfnsuIAJM3ekGPzZmnFCqurAQAAQEql6BTE6dOnO/0cHBzs9LNhGAoPD9eaNWvUu3dv11UHIEnly5vXg02aZDbkePJJGnIAAABkBmnqgrh3716nnz08PFS4cGHNmDHjnh0SAbjG6NHSJ59IJ05Ib74pTZlidUUAAAC4lzQFsC1btri6DgCpZG/I0a7drYYclSpZXRUAAADuJk1t6I8fP66jR48mmn/06FGdOHHifmsCkEJt25r3A4uNpSEHAABAZpCmANanTx9t37490fxffvlFffr0ud+aAKSQzSbNmWM25Ni0SfryS6srAgAAwN2kKYDt3btXTzzxRKL5jz32mPbt23e/NQFIhfLlpZEjzccjRkjR0dbWAwAAgOSlKYDZbLYk7wd25coVxcfH33dRAFJn9GipTBnpr7/MhhwAAABwT2kKYA0aNNDkyZOdwlZ8fLwmT56sevXquaw4ACnj5yfNmmU+njFDOnLE0nIAAACQjDR1QZw6daoaNGigihUrqn79+pKkbdu2KSoqSt99951LCwSQMk8/bd4PbN06syHHhg3mNWIAAABwH2k6Ala5cmX997//1bPPPqvz58/r6tWr6tWrlw4fPqwqVaq4ukYAKWBvyJEjhxQWJq1caXVFAAAAuFOajoBJUvHixfX222+7shYA9+nBB82GHG++aTbkaNVKypXL6qoAAABgl6YjYADc15gx0gMPSKdPS2+9ZXU1AAAAuB0BDMhibm/IMX269L//WVoOAAAAbkMAA7Kgdu3M0w9jY82GHIZhdUUAAACQCGBAlnR7Q46NG6VVq6yuCAAAABIBDMiyKlSQXn3VfDxihBQdbW09AAAASEUXxOrVq2vz5s3Knz+/Hn30UdnucoOhPXv2uKQ4APfntdekTz+VTp2S3n6bphwAAABWS3EAa9eunXx8fCRJ7du3T696ALiQvSFHx45mQ47evaV//cvqqgAAALKvFAewkJCQJB8DcG/t20stW0obNkgvvSStW2deIwYAAICMxzVgQBZns0lz55oNOTZskFavtroiAACA7CtNASw+Pl7Tp09X7dq1FRAQoAIFCjhNANxLhQrSK6+Yj4cPl65ft7QcAACAbCtNAWzixImaOXOmunTpoitXrig4OFgdO3aUh4eHJkyY4OISAbjCa69JpUrdasgBAACAjJemAPbZZ5/pww8/1MsvvywvLy9169ZN//73vzV+/Hj9/PPPrq4RgAvkymU25JCkd96Rjh61tBwAAIBsKU0BLCIiQlWrVpUk5c6dW1euXJEkPfXUU1qzZo3rqgPgUh06SC1aSDEx0rBhkmFYXREAAED2kqYAVrJkSYWHh0uSypcvr40bN0qSdu7c6WhVD8D92BtyeHub3RC/+srqigAAALKXNAWwDh06aPPmzZKkoUOHaty4capQoYJ69eqlfv36ubRAAK71r3/RkAMAAMAqKb4P2O2mTJnieNylSxeVLl1aO3bsUIUKFdS2bVuXFQcgfbz+uvSf/0gnT0qTJ0tvvGF1RQAAANlDmgLYnerWrau6deu6YlEAMkCuXNK770rPPCNNmyb17i09+KDVVQEAAGR9aToFcfHixU7NNkaOHKl8+fLp8ccf18mTJ11WHID007Gj1Ly52ZDjpZdoyAEAAJAR0hTA3n77beXMmVOStGPHDoWGhmratGkqVKiQRowY4dICAaSPOxtyfP211RUBAABkfWkKYKdPn9aD/3++0urVq/XMM89o4MCBmjx5srZt2+bSAgGkn4oVpZdfNh8PG0ZDDgAAgPSWpgCWO3duXbp0SZK0ceNGNW/eXJLk6+urf/75x3XVAUh3Y8dKJUuaDTlu668DAACAdJCmANa8eXM9//zzev755/W///1PrVu3liT9/vvvKlOmjCvrA5DO7A05JLMhx7Fj1tYDAACQlaUpgM2bN09169bVhQsX9OWXX6pgwYKSpN27d6tbt24uLRBA+uvUSWrWTLp5k4YcAAAA6SlNbejz5cun0NDQRPMnTpx43wUByHj2hhyPPCKtXSt984309NNWVwUAAJD1pOkIGICsp1IlKTjYfDxsmMTlnAAAAK5HAAPgYG/IceIEDTkAAADSAwEMgEPu3NLMmebjqVNpyAEAAOBqBDAATp55Rmra1GzIMXy41dUAAABkLQQwAE5sNik0VPL2lr791mzIAQAAANdIUxfERx99VDabLdF8m80mX19fPfjgg+rTp48aN2583wUCyHiVKkkjRpj3BRs2zGxRnzOn1VUBAABkfmk6AtaqVSv9+eefypUrlxo3bqzGjRsrd+7cOnbsmGrVqqXw8HA1a9ZMX331VYqWN2/ePJUpU0a+vr6qU6eOfv3117uOX758uSpVqiRfX19VrVpVa9eudXp+5cqVatGihQoWLCibzaZ9+/YlWsaNGzc0ZMgQFSxYULlz51anTp107ty5FG8DIKsbN04qUUI6fty8HgwAAAD3L00B7OLFi3r55Ze1bds2zZgxQzNmzNAPP/ygV155RdHR0dq4caPGjh2rN954457LWrZsmYKDgxUSEqI9e/aoWrVqatmypc6fP5/k+O3bt6tbt27q37+/9u7dq/bt26t9+/Y6cOCAY0x0dLTq1aunqXf5rXHEiBH65ptvtHz5cn3//fc6e/asOnbsmPqNAWRRtzfkmDJF+vNPa+sBAADICtJ0CuIXX3yh3bt3J5rftWtX1ahRQx9++KG6deummfbf3u5i5syZGjBggPr27StJmj9/vtasWaOFCxdq9OjRicbPnj1brVq10quvvipJeuONNxQWFqbQ0FDNnz9fktSzZ09J0okTJ5Jc55UrV/TRRx9pyZIlatKkiSTp448/1kMPPaSff/5Zjz32WJKvu3nzpm7evOn4OSoqSpIUGxur2NjYe77X9GRfv9V1IGtp315q3NhTW7Z46KWXErRqVbzVJWUIvk/ZC/s7fbF9nbE93Fdm2DfuVqM71OMONdiltIY0BTBfX19t375dDz74oNP87du3y9fXV5KUkJDgeJycmJgY7d69W2PGjHHM8/DwULNmzbRjx44kX7Njxw4F2+8W+/9atmyp1atXp7j+3bt3KzY2Vs2aNXPMq1SpkkqXLq0dO3YkG8AmT56siRMnJpq/ceNG+fn5pXj96SksLMzqEpDFdOyYWz/80Fhr1nho4sRfVatW9jlVl+9T9sL+Tl9sX2dsD/eVGfaNu9XoDvW4Qw3Xr19P0bg0BbChQ4dq0KBB2r17t2rVqiVJ2rlzp/7973/rtddekyRt2LBBgYGBd13OxYsXFR8fr6JFizrNL1q0qA4fPpzkayIiIpIcHxERkeL6IyIilCNHDuXLly9VyxkzZoxT+IuKilKpUqXUokUL+fv7p3j96SE2NlZhYWFq3ry5vL29La0FWc/x44ZmzpSWLKmjkSPjsnxDDr5P2Qv7O32xfZ2xPdxXZtg37lajO9TjDjXY2c+Ou5c0BbCxY8eqbNmyCg0N1aeffipJqlixoj788EN1795dkjRo0CANHjw4LYt3Wz4+PvLx8Uk039vb2/IdbudOtSDrmDBBWrpUOn7cpnff9VZIiNUVZQy+T9kL+zt9sX2dsT3cV2bYN+5WozvU4y41pESaApgkPffcc3ruueeSfT5nCv6LvFChQvL09EzUffDcuXMKCAhI8jUBAQGpGp/cMmJiYhQZGel0FCy1ywGyizx5zIYcXbuaDTl69ZLKlrW6KgAAgMznvm7EHBMTo7/++kunTp1ymlIqR44cqlGjhjZv3uyYl5CQoM2bN6tu3bpJvqZu3bpO4yXznM/kxielRo0a8vb2dlrOkSNHdOrUqVQtB8hOnn1WatJEunFDGj7c6moAAAAypzQdATt69Kj69eun7du3O803DEM2m03x8SnvlBYcHKzevXurZs2aql27tmbNmqXo6GhHV8RevXqpRIkSmjx5siRp2LBhatiwoWbMmKE2bdpo6dKl2rVrlxYsWOBY5uXLl3Xq1CmdPXtWkhmuJPPIV0BAgPLmzav+/fsrODhYBQoUkL+/v4YOHaq6desm24ADyO5sNmnuXKlaNenrr6U1a6Q2bayuCgAAIHNJUwDr06ePvLy89O2336pYsWKy2WxpLqBLly66cOGCxo8fr4iICAUGBmr9+vWORhunTp2Sh8etA3WPP/64lixZorFjx+q1115ThQoVtHr1alWpUsUx5uuvv3YEOMlsjy9JISEhmjBhgiTp3XfflYeHhzp16qSbN2+qZcuWeu+999L8PoDsoHJl8+jX9OnSSy9JTZtK92h2CgAAgNukKYDt27dPu3fvVqVKlVxSRFBQkIKCgpJ8buvWrYnmde7cWZ07d052eX369FGfPn3uuk5fX1/NmzdP8+bNS02pQLY3fry0ZIl5Y+Zp08yfAQAAkDJpugascuXKunjxoqtrAZAJ5MkjzZhhPp48WTp+3Np6AAAAMpM0BbCpU6dq5MiR2rp1qy5duqSoqCinCUDW1qWL1Lix2ZBjxAirqwEAAMg80nQKYrNmzSRJTZs2dZqfliYcADIfe0OOwEDpq6+ktWul1q2trgoAAMD9pSmAbdmyxdV1AMhkHn5YGjbMPB3xpZfMFvU05AAAALi7VAewNWvW6MaNG2rQoIEKFy6cHjUByCRCQsyGHMeOmZ0Rx461uiIAAAD3luoAtnbtWl27dk0DBgzQ66+/rkOHDkmSHn74YfXr10958+Z1eZEA3JO9IUf37tJbb0k9ekhlylhdFQAAgPtKdROOefPmqWXLloqMjNS7776ry5cv6/Lly5o5c6bKly+vPXv2pEedANxU165Sw4Y05AAAAEiJNHVB7NOnj1q2bKkTJ05o5cqVWrlypY4fP66nnnpKw4cPd3GJANyZzSbNmyd5ekqrV0vr1lldEQAAgPtKUwDz9PTUu+++Ky+vW2cwenl5aeTIkdq1a5fLigOQOdgbckhmQ46bN62tBwAAwF2lKYD5+/vr1KlTieafPn1aefLkue+iAGQ+ISFSQID0xx9mQw4AAAAklqYA1qVLF/Xv31/Lli3T6dOndfr0aS1dulTPP/+8unXr5uoaAWQC/v5mQw7JbMhx8qS19QAAALijNN0HbPr06bLZbOrVq5fi4uIkSd7e3ho8eLCmTJni0gIBZB7dukkLFkjff2825Fi50uqKAAAA3EuqA9jw4cMVFxenTp06afLkyfrjjz9ks9lUvnx5+fn5pUeNADIJm00KDZUCA6VVq6T166VWrayuCgAAwH2k+hTE+vXrq3r16urSpYv8/Pz0yCOPqGrVqoQvAJKkKlXMRhySNHQoDTkAAABul+ojYJ06dZKHh4dsNpuaNGmS5JjvvvvuvgsDkHlNmCB9/rnZkGPGDOm116yuCAAAwD2kqQlH+/bt9fzzz6tatWqqVq2aKleurJiYGO3evVtVq1Z1dY0AMhl//1udEN98k4YcAAAAdmlqwrEymSvrJ0yYoGvXrt1XQQCyhu7dzYYcP/wgBQdLX35pdUUAAADWS9MRsOT06NFDCxcudOUiAWRS9oYcnp5mN8QNG6yuCAAAwHouDWA7duyQr6+vKxcJIBOrWtVsxCHRkAMAAEBK4ymIHTt2dPrZMAyFh4dr165dGjdunEsKA5A12BtyHD0qzZwpjRljdUUAAADWSdMRsLx58zpNBQoUUKNGjbR27VqFhIS4ukYAmVjevLcacrzxhnTqlLX1AAAAWClNR8A+/vhjV9cBIAt77jmzIce2bWZDjhUrrK4IAADAGi69BgwAknJ7Q44vv5Q2brS6IgAAAGukKYCdO3dOPXv2VPHixeXl5SVPT0+nCQDu9MgjUlCQ+ZiGHAAAILtK0ymIffr00alTpzRu3DgVK1ZMNpvN1XUByIImTpSWLpX+9z/p3Xel0aOtrggAACBjpSmA/fjjj9q2bZsCAwNdXA6ArCxvXumdd6RevcyGHN27S6VLW10VAABAxknTKYilSpWSYRiurgVANtCjh1SvnnT9uvTyy1ZXAwAAkLHSFMBmzZql0aNH68SJEy4uB0BWZ2/I4eFhdkMMC7O6IgAAgIyT4lMQ8+fP73StV3R0tMqXLy8/Pz95e3s7jb18+bLrKgSQ5VSrZjbkmDPHbMjx3/9KOXJYXRUAAED6S3EAmzVrVjqWASC7sTfkOHLEbMgxapTVFQEAAKS/FAew3r17p2cdALKZfPnMhhy9e0uTJpkNOUqVsroqAACA9JWqa8ASEhI0depUPfHEE6pVq5ZGjx6tf/75J71qA5DF9ewpPfEEDTkAAED2kaoA9tZbb+m1115T7ty5VaJECc2ePVtDhgxJr9oAZHE2mzRvntmQY/lyadMmqysCAABIX6kKYJ988onee+89bdiwQatXr9Y333yjzz77TAkJCelVH4Asrlo1yf7/OEFBUkyMtfUAAACkp1QFsFOnTql169aOn5s1ayabzaazZ8+6vDAA2cekSVKRImZDDvr9AACArCxVASwuLk6+vr5O87y9vRUbG+vSogBkL/nySdOmmY8nTZL++svScgAAANJNirsgSpJhGOrTp498fHwc827cuKFBgwYpV65cjnkrV650XYUAsoWePaUFC6Tt282GHMuWWV0RAACA66UqgCXVir5Hjx4uKwZA9uXhYTbkqFFD+uILacAAqVkzq6sCAABwrVQFsI8//ji96gAABQZKL74ohYZKQ4dK+/dLOXJYXRUAAIDrpOoaMABIb2+8IRUuLB0+LM2ebXU1AAAArkUAA+BWbm/IMXEiDTkAAEDWQgAD4HZ69ZLq1pWio6VXXrG6GgAAANchgAFwO/aGHB4eZjfE776zuiIAAADXIIABcEuPPioNHmw+DgqSYmKsrQcAAMAVCGAA3Ja9IcehQ9KcOVZXAwAAcP8IYADcVv780tSp5uMJE6QzZywtBwAA4L4RwAC4td69pcceoyEHAADIGghgANyavSGHzSYtXSpt2WJ1RQAAAGlHAAPg9qpXd27IERtrbT0AAABpRQADkCm8+aZUqJB08CANOQAAQOZFAAOQKdzZkOPsWUvLAQAASBMCGIBMo08fqU4d6do1GnIAAIDMiQAGINO4vSHH559LW7daXREAAEDqEMAAZCo1akiDBpmPhwyhIQcAAMhcCGAAMp0335QKFjQbcsyda3U1AAAAKUcAA5DpFCggTZliPg4JoSEHAADIPAhgADKlfv2k2rXNhhyvvmp1NQAAAClDAAOQKd3ekGPJEun7762uCAAA4N4IYAAyrZo1pRdeMB/TkAMAAGQGBDAAmdpbb5kNOX7/XQoNtboaAACAuyOAAcjUChSQJk82H4eESOHh1tYDAABwNwQwAJle//5mQ46rV2nIAQAA3BsBDECmd3tDjs8+k374weqKAAAAkkYAA5Al1KwpDRxoPqYhBwAAcFcEMABZxltvmdeEHThgHhEDAABwNwQwAFlGwYK3GnKMH09DDgAA4H4IYACylP79pVq1zIYcI0daXQ0AAIAzAhiALMXT81ZDjv/8h4YcAADAvRDAAGQ5tWpJAwaYj4OCpLg4a+sBAACwI4AByJLefttsyPHbbzTkAAAA7oMABiBLKljQDGGS2ZAjIsLaegAAACQCGIAs7PnnzfuDRUXRkAMAALgHAhiALOv2hhyffipt22Z1RQAAILsjgAHI0mrXNo+ESdKQITTkAAAA1nKbADZv3jyVKVNGvr6+qlOnjn799de7jl++fLkqVaokX19fVa1aVWvXrnV63jAMjR8/XsWKFVPOnDnVrFkzHT161GlMmTJlZLPZnKYpU6a4/L0BsNbtDTnee8/qagAAQHbmFgFs2bJlCg4OVkhIiPbs2aNq1aqpZcuWOn/+fJLjt2/frm7duql///7au3ev2rdvr/bt2+vAgQOOMdOmTdOcOXM0f/58/fLLL8qVK5datmypGzduOC1r0qRJCg8Pd0xDhw5N1/cKIOMVKiS99Zb5eNw4GnIAAADreFldgCTNnDlTAwYMUN++fSVJ8+fP15o1a7Rw4UKNHj060fjZs2erVatWevXVVyVJb7zxhsLCwhQaGqr58+fLMAzNmjVLY8eOVbt27SRJn3zyiYoWLarVq1era9eujmXlyZNHAQEBKarz5s2bunnzpuPnqKgoSVJsbKxiY2PT9uZdxL5+q+sA3FWfPtKHH3pqzx4PvfpqghYujE92LN+n7IX9nb7Yvs7YHu4rM+wbd6vRHepxhxrsUlqDzTAMI51ruauYmBj5+flpxYoVat++vWN+7969FRkZqa+++irRa0qXLq3g4GANHz7cMS8kJESrV6/W/v379eeff6p8+fLau3evAgMDHWMaNmyowMBAzZ49W5J5CuKNGzcUGxur0qVLq3v37hoxYoS8vJLOpRMmTNDEiRMTzV+yZIn8/PzStgEAZJj//S+/Ro5sIEl6++1tqlz5ssUVAQCArOL69evq3r27rly5In9//2THWX4E7OLFi4qPj1fRokWd5hctWlSHDx9O8jURERFJjo/4//OK7H/ebYwkvfTSS6pevboKFCig7du3a8yYMQoPD9fMmTOTXO+YMWMUHBzs+DkqKkqlSpVSixYt7rqRM0JsbKzCwsLUvHlzeXt7W1oL4K5at5YOH07QwoUe+vzzevrllzgl9f8tfJ+yF/Z3+mL7OmN7uK/MsG/crUZ3qMcdarCznx13L5YHMCvdHqYeeeQR5ciRQy+88IImT54sHx+fRON9fHySnO/t7W35Drdzp1oAdzR1qrRqlfTbbzb9+9/euttln3yfshf2d/pi+zpje7ivzLBv3K1Gd6jHXWpICcubcBQqVEienp46d+6c0/xz584le21WQEDAXcfb/0zNMiWpTp06iouL04kTJ1L7NgBkEnc25LjjrwkAAIB0ZXkAy5Ejh2rUqKHNmzc75iUkJGjz5s2qW7dukq+pW7eu03hJCgsLc4wvW7asAgICnMZERUXpl19+SXaZkrRv3z55eHioSJEi9/OWALi5gQOl6tWlK1ekUaOsrgYAAGQnbnEKYnBwsHr37q2aNWuqdu3amjVrlqKjox1dEXv16qUSJUpo8uTJkqRhw4apYcOGmjFjhtq0aaOlS5dq165dWrBggSTJZrNp+PDhevPNN1WhQgWVLVtW48aNU/HixR2NPnbs2KFffvlFjRs3Vp48ebRjxw6NGDFCPXr0UP78+S3ZDgAyhqenNG+eVLeutHixNGCA9MQTVlcFAACyA7cIYF26dNGFCxc0fvx4RUREKDAwUOvXr3c00Th16pQ8PG4drHv88ce1ZMkSjR07Vq+99poqVKig1atXq0qVKo4xI0eOVHR0tAYOHKjIyEjVq1dP69evl6+vryTzeq6lS5dqwoQJunnzpsqWLasRI0Y4XRcGIOt67DGpf3/po4+kIUOkXbuUZEMOAAAAV3KbXzeCgoIUFBSU5HNbt25NNK9z587q3Llzssuz2WyaNGmSJk2alOTz1atX188//5ymWgFkDZMnS19+Ke3fL82fLyXzVxAAAIDLWH4NGABYpXDhWw05xo6Vzp+3th4AAJD1EcAAZGsvvCA9+igNOQAAQMYggAHI1uwNOSRp0SJp+3ZLywEAAFkcAQxAtle3rtSvn/n4xRelLVts+uGHEvr+e5vi462tDQAAZC0EMACQNGWK5OdnNuRo2dJLM2fWVPPmXipTRlq50urqAABAVkEAAwBJ27ZJ168nnn/mjPTMM4QwAADgGgQwANlefLw0bFjSzxmG+efw4eJ0RAAAcN/c5j5gAGCVbdukv/5K/nnDkE6flurVM2/gXKGC9OCD5p+lS5uNPAAAAFKCAAYg2wsPT9m4n382p9t5e0vlyt0KZLf/Wbq05MXfsgAA4Db8agAg2ytWLGXjRoyQPDyko0elP/6Qjh2Tbt6Ujhwxpzt5e0tlyyYOZvYjZ4QzAACyH/75B5Dt1a8vlSxpNtywX/N1O5vNfP6dd5xPN4yPN19jD2S3/2kPZ//7nzndyR7Obg9m9scPPEA4AwAgq+KfeADZnqenNHu22e3QZnMOYTab+eesWYmv9fL0NI9klS4tNW3q/FxCgnld2Z3BzH7k7MaN5MOZl1ficGb/s0wZwhkAAJkZ/4wDgKSOHaUVK8xuiLc35ChZ0gxfHTumbnkeHrfCWZMmzs8lJDgfObs9nP3xhxnOjh41p3XrnF/r5WWGsKROa3zgAfPIGgAAcF8EMAD4fx07Su3aSVu2xGndun168slANW7s5fIuhx4eUqlS5pRUODt7NunTGu3hzB7U7uTpefcjZ4QzAACsRwADgNt4ekoNGxqKjj6jhg2rZXiLeQ8P86hbyZJS48bOz9nDWVLB7I8/pH/+ufV4/Xrn13p6miEsqXBWtizhDACAjEIAA4BM4vZw1qiR83MJCWY7/eSOnP3zj3nt2bFj0oYNzq/19DRPX0zqtMYyZaQcOTLqHQIAkPURwAAgC/DwkEqUMKc7w5lh3DpyllQ4u35d+vNPc0ounCV35IxwBgBA6hDAACCLs9luhbOGDZ2fMwzzyFlypzVGR98KZxs3Or/WwyP5I2eEMwAAkkYAA4BszGaTihc3pwYNnJ8zDCkiIvnTGqOjpePHzSmpcFa6dPLhzMcn494jAADuhAAGAEiSzSYVK2ZOyYWz5E5rvHZNOnHCnMLCnF9rD2dJndZYrhzhDACQtRHAAACpdns4q1/f+TnDkM6dSzqYHT3qHM42bUq83LsdOfP1de37iI+Xvv/eph9+KKFcuWxq3DjxDbcBAHAlAhgAwKVsNikgwJzq1XN+zjCk8+eTP63x6lXp5ElzSi6c3R7M7I/LlUt9OFu50n7jbS9JNTVzptlhcvbs1N94GwCAlCKAAQAyjM0mFS1qTsmFs+SOnN0ezjZvTrzcUqWSPq2xfPnE4WzlSumZZ8x13u7MGXP+ihWEMABA+iCAAQDcwu3h7IknnJ8zDOnCheTDWVSUdOqUOX33XeLllizpHMimTUscvuzrsdmk4cOldu04HREA4HoEMACA27PZpCJFzOnxx52fMwzp4sWkg5k9nJ0+bU53hrOkGIY5dtu2xPdUAwDgfhHAAACZms0mFS5sTsmFs9uD2aZN0o4d917uN99INWpIefKkT90AgOyJAAYAyLJuD2d165rzmjSRGje+92tnzpTmzjVPh3zySalVK6lqVXOZAACklYfVBQAAkJHq1zevCbtbkMqd27xWLDZW2rpVGjVKqlZNKlFC6tdP+uIL6e+/M6xkAEAWQgADAGQrnp5mq3kpcQiz2cxp8eJbpy3OnSu1aSP5+Unh4dLHH0tdukiFCplHx954Q9q5U0pIyPj3AgDIfAhgAIBsp2NHs9V8iRLO80uWdG5B/+CDUlCQ9O230qVLUliY9PLL0sMPm4Fr+3Zp/Hipdm2ze+Nzz0n/+Y/ZTh8AgKQQwAAA2VLHjtKJE1JYWJyCg3cpLCxOx48nf/8vX1+pWTNp+nTpwAHzfmQLFpjj/f3NZh9Llkg9e5phrGZNaexY6ccfpbi4DH1rAAA3RgADAGRbnp5Sw4aGGjQ4o4YNjVTd96t0aWnAAOnLL83w9f330pgx0qOPms/v3i299ZZ5zVmhQlLnztJHH0l//ZU+7wUAkDnQBREAgPvk7S01aGBOb78tRURIGzdK69dLGzZIly+bpzauWGGOr1LlVmfFJ56QfHysrR8AkHEIYAAAuFhAgNSrlznFx0u7dplhbP166ZdfzFMYDxyQ3nlHypXLbI3fqpU5lStndfUAgPREAAMAIB15ekp16phTSMitZh72QHbunHnT52++Mcf/61+3wljDhmb3RQBA1kEAAwAgAxUsKHXtak4JCdJ//3srjP30k/S//5nTnDnmqYkNG5ph7MknpYoVuRE0AGR2NOEAAMAiHh5SYKA0erR5w+eLF6WVK6WBA6VSpaSbN81ryYKDpYceksqWlQYNklavlqKiLC4eAJAmHAEDAMBN5M0rdehgToYhHT5sHhlbt87ssnjypPTBB+bk5SXVq3frdMVHHuHoGABkBhwBAwDADdls5lGvESPMo2CXL0tr1khDh0oVKpj3Ftu61Tx6Fhho3lS6b19p2TJzLADAPXEEDACATCBXLql1a3OSpGPHbl079t13Uni4tGiROXl4mE0/7EfHatRQqu5xBgBIPxwBAwAgEypfXhoyxOyeePmytGmT9Mor0sMPm809duwwuy7WqSMVLSp17y59+qnZdREAYB0CGAAAmZyPj9S0qXlfsQMHpFOnpA8/lDp1kvz9zdb3n39u3pcsIMA8Ivb669KPP5qnMgIAMg4BDACALKZUKen556UVK8zOitu2Sa+9JlWvbj6/Z4/09ttS/fpSoULSM89I//639Ndf1tYNANkB14ABAJCFeXub3RLr1ZPeess8BXHjRvPasQ0bzKNjX35pTpJUpcqta8fq1TOPrgEAXIcABgBANlK0qNSzpznFx0u7d99q5vHLL+YpjAcOSNOnS35+UpMmt24EXa6c1dUDQOZHAAMAIJvy9JRq1zan8eNvNfNYt84MZBER0rffmpNktr+3Hx1r1MgMaHcTHy99/71NP/xQQrly2dS4Md0YAYBrwAAAgCSpQAHp2Weljz+Wzp6V9u2Tpkwxw5aXl3T0qDR3rtSmjTm2RQvp3XelQ4fMG0ffbuVKqUwZqXlzL82cWVPNm3upTBlzPgBkZxwBAwAAidhsUrVq5jRqlBQVZd5vbP168wjZqVNSWJg5BQdLpUvfOjoWHW12XLwzlJ05Yzb8WLFC6tjRmvcFAFYjgAEAgHvy95fatzcnw5COHLkVxr7/3gxkCxaYU3IMwwx2w4dL7dpxOiKA7IlTEAEAQKrYbFKlSmaQ2rDBvHZs7VrppZekkiXv/lrDkE6flubNM09zvPMoGQBkdRwBAwAA98XPz+yS+OST0mOPSd273/s1w4aZU86cUvnyztODD5p/PvCAee0ZAGQl/LUGAABcplixlI0LCJDOn5f++edW6/s7eXqajTySCmflyt27CyMAuCMCGAAAcJn69c3TEM+cSfr0QpvNfP74cbNN/cmT0rFjt6Y//jD//PNP6caNW/OTUqzYrUB2ezgrX97s0ggA7ogABgAAXMbTU5o92+x2aLM5hzCbzfxz1ixznKeneW+xChUSLychwbxGLKlwduyYFBkphYeb07ZtiV+fL1/y4axYMcmDq+ABWIQABgAAXKpjR7PV/LBh0l9/3ZpfsqQZvlLSgt7DwxxfsqTUsGHi5y9fdg5ktwe08HAzoO3aZU538vW9+3Vn3t5pfed3x42pAUgEMAAAkA46djRbzW/ZEqd16/bpyScD1bixl8sCR4ECUu3a5nSn6GjzFMakwtnJk+apjb//bk538vQ0Q1hy153lypW2eleutAdSL0k1NXOmGS5nz+aeaEB2QwADAADpwtNTatjQUHT0GTVsWC3DjvbkyiVVrWpOd4qNNe9ZdvvRs9uvO/vnH/PPP/80bzJ9p4CAu193Zj/N8nYrV5qnZHJjagASAQwAAGQj3t63AtOdDMM8fTG5Uxv//luKiDCnH39M/Pq8eZ0DWfnyUtmyUlBQ0g1JuDE1kD0RwAAAAGSGoeLFzalBg8TP//134mYg9sdnz0pXrki7d5tTStlvTP3ll2YI8/Fx3fsB4J4IYAAAACmQP79Us6Y53en6dbO1/p1Hz/buNe93di9duph/5ssnFS1qTkWK3Hqc1Ly0Xo8GwFoEMAAAgPvk5yc9/LA53W7rVqlx43u/3sPDbL0fGWlOR47c+zW5cqUsqBUtap4emdT1aQAyHgEMAAAgnaT0xtTHjklXr0rnzjlP588nnnfunNnJ0d7t8c8/711HjhwpC2pFi0oFC3KfNCA9EcAAAADSSUpvTO3tbXZRLFBAeuihuy/TMKRr1+4d1OzzoqKkmBjznmy335ctOR4eUuHC9w5qRYqYU3rdNw3IqghgAAAA6cgVN6a+nc0m5cljTg8+eO/x//xzK4zdLaidOyddumSeCmn/OSUKFEj5dWs5c6buvQJZEQEMAAAgnaX3janvJmdO8+bSDzxw77GxsdKFC/cOaufOmePi46XLl83p0KF7Lz9PnuSD2p3z8+TJuOvW4uOl77+36YcfSihXLpsaN+a2AEg/BDAAAIAMYNWNqVPD2/tWK/57SUgwg1dKr1uLiTGvc7t61ewWeS++vilvMpI/f9qvW1u50n500ktSTc2caR6dnD2bG2QjfRDAAAAAkGoeHlKhQuZ0Z/fHOxmGeS1aSsLa+fPmNW43bkgnT5rTvXh5Jb5uLbnwVqiQOV4yw9czzyRukHLmjDl/xQpCGFyPAAYAAIB0ZbOZrfDz5pX+9a97j79+PeVNRv7+W4qLk8LDzSkltRQsaIazY8eS7k5pGOa44cPNU0fd8Wil1ThtM+0IYAAAAHArfn5S2bLmdC8xMSlvMnLxonnq5MWL5nQ3hiGdPi0FBJh1FCvmPBUvfutxkSK3jqqlF3cKPJy2eX8IYAAAAMi0cuQwf/kvWfLeY+PjzeB1/rz0+efS5Mn3fk1KwpqHhxnCkgto9p+LFjXrTS13CjzudNqmO4XS1CCAAQAAIFvw9Lx1LdilSykLYO+/b4an8HDp7Nlbpzrafz53zjyqFhFhTnv33n15hQolHdDunOfra453t8AzbJh7nLbpTqE0tQhgAAAAyHbq1zd/YT9zJulAYbOZzw8YcPcwER9vtuRPKqDdPi8iwmzzbz+i9ttvd68vXz4ziP35Z/KBR5Kef95cvmGYQTA+3vzz9sd3m5ea8RERd7+Zt/20zf79pSpVzFNJk5py5Uo8LzU39HanUJoWBDAAAABkO56e5tGSZ54xw9btv8zb7z82a9a9j+R4eprXiQUESI8+mvw4e9v+pI6i3fnzzZtSZKQ53cvff0tBQfcel5EWL079a7y8kg9nt8/39ZX+8x/3OAqXVm4TwObNm6d33nlHERERqlatmubOnavatWsnO3758uUaN26cTpw4oQoVKmjq1Klq3bq143nDMBQSEqIPP/xQkZGReuKJJ/T++++rQoUKjjGXL1/W0KFD9c0338jDw0OdOnXS7NmzlTt37nR9rwAAALBex47m0RLzVLZb80uWNMOXK4+i3N62/5FHkh9nGGbwCg+XPvtMevvtey+7Vi3zRtuenuZ67vzzfubd/tyff0qhofeup21b8wje9evOU3R04p/tQSouzrxVQVRUSrZm8uxH4bZtkxo1ur9lpRe3CGDLli1TcHCw5s+frzp16mjWrFlq2bKljhw5oiJFiiQav337dnXr1k2TJ0/WU089pSVLlqh9+/bas2ePqlSpIkmaNm2a5syZo8WLF6ts2bIaN26cWrZsqYMHD8r3/0+qfe655xQeHq6wsDDFxsaqb9++GjhwoJYsWZKh7x8AAADW6NjRPFqyZUuc1q3bpyefDFTjxl6WHT2x2cwbS+fPLzVvnrIANm1axoSN+Hhp9ep7n7a5alXKjj4ZhtnFMqlwllx4+/VX8xTEe0nJLQksY7iB2rVrG0OGDHH8HB8fbxQvXtyYPHlykuOfffZZo02bNk7z6tSpY7zwwguGYRhGQkKCERAQYLzzzjuO5yMjIw0fHx/j888/NwzDMA4ePGhIMnbu3OkYs27dOsNmsxlnzpxJUd1XrlwxJBlXrlxJ2RtNRzExMcbq1auNmJgYq0sBMj2+T9kL+zt9sX2dsT3clzvum7g4wyhZ0jBsNsMw44rzZLMZRqlS5riM8uWX5nrvrMk+78sv03f9W7YkvS3unLZsSd86kpLSbGD5EbCYmBjt3r1bY8aMcczz8PBQs2bNtGPHjiRfs2PHDgUHBzvNa9mypVavXi1JOn78uCIiItSsWTPH83nz5lWdOnW0Y8cOde3aVTt27FC+fPlUs2ZNx5hmzZrJw8NDv/zyizp06JBovTdv3tTNmzcdP0f9/zHS2NhYxcbGpv7Nu5B9/VbXAWQFfJ+yF/Z3+mL7OmN7uC933TczZtjUtavn/1+nZnPMt9nMQ1DTp8crIcFQQkLG1NO2rbR0qU3BwZ46c+ZWPSVKGJoxI15t2xpKz0342GNSiRJeOnvWeXvY2WyGSpSQHnssLl3rSEpKPzuWB7CLFy8qPj5eRYsWdZpftGhRHT58OMnXREREJDk+IiLC8bx93t3G3Hl6o5eXlwoUKOAYc6fJkydr4sSJieZv3LhRfn5+yb3FDBUWFmZ1CUCWwfcpe2F/py+2rzO2h/tyt33j4yONHFlM//53VV26lNMxv2DBf9S//wH5+IRr7dqMr2nOHOngwYL6+29f5c9/Q5UrX5KnpzKklh49imnq1FqSDEm3hzBDhiE999xObdiQ8ecgXr9+PUXjLA9gmcmYMWOcjrxFRUWpVKlSatGihfz9/S2szEzcYWFhat68ubxT08cTQCJ8n7IX9nf6Yvs6Y3u4L3feN61bSxMmSFu33lBY2AE1b15FjRp5y9PzUUl3ab2Yzlq1smabtW4tVa8e//9H4W7NL1lSmjEjXh06WLNdolLYQcTyAFaoUCF5enrq3LlzTvPPnTungICAJF8TEBBw1/H2P8+dO6dixYo5jQkMDHSMOX/+vNMy4uLidPny5WTX6+PjIx8fn0Tzvb293eaL6k61AJkd36fshf2dvti+ztge7std9423t9S0qXTz5hk1bVrNrWq0Yps9+6zUqVNSzVOsizcp3QYe6VzHPeXIkUM1atTQ5s2bHfMSEhK0efNm1a1bN8nX1K1b12m8ZB4uto8vW7asAgICnMZERUXpl19+cYypW7euIiMjtXv3bseY7777TgkJCapTp47L3h8AAAAA1/P0lBo2NNSgwRk1bGi47X2/7mT5ETBJCg4OVu/evVWzZk3Vrl1bs2bNUnR0tPr27StJ6tWrl0qUKKHJkydLkoYNG6aGDRtqxowZatOmjZYuXapdu3ZpwYIFkiSbzabhw4frzTffVIUKFRxt6IsXL6727dtLkh566CG1atVKAwYM0Pz58xUbG6ugoCB17dpVxYsXt2Q7AAAAAMja3CKAdenSRRcuXND48eMVERGhwMBArV+/3tFE49SpU/LwuHWw7vHHH9eSJUs0duxYvfbaa6pQoYJWr17tuAeYJI0cOVLR0dEaOHCgIiMjVa9ePa1fv95xDzBJ+uyzzxQUFKSmTZs6bsQ8Z86cjHvjAAAAALIVtwhgkhQUFKSgoKAkn9u6dWuieZ07d1bnzp2TXZ7NZtOkSZM0adKkZMcUKFCAmy4DAAAAyDCWXwMGAAAAANkFAQwAAAAAMggBDAAAAAAyCAEMAAAAADIIAQwAAAAAMggBDAAAAAAyCAEMAAAAADIIAQwAAAAAMggBDAAAAAAyCAEMAAAAADKIl9UFZGaGYUiSoqKiLK5Eio2N1fXr1xUVFSVvb2+rywEyNb5P2Qv7O32xfZ2xPdxXZtg37lajO9TjDjXY2TOBPSMkhwB2H65evSpJKlWqlMWVAAAAAHAHV69eVd68eZN93mbcK6IhWQkJCTp79qyaNGmiXbt23ffyatWqpZ07d6bptVFRUSpVqpROnz4tf3//+64F6ed+9nNmlpnetzt9nzJ6u6X3+ly9fFcsz532d1bkTtvXHf4esmJ7uMP7Tg2r6r1z37jj37/38/lJj79/N2/efN+f5/uty53+jjEMQ1evXlXx4sXl4ZH8lV4cAbsPHh4eKlmypLy8vFyywz09Pe97Of7+/pZ/+HB3rtjPmVFmfN/u8H3K6O2W3utz9fJduTx32N9ZmTtsX3f6eygjt4c7ve+UsLpe+75x579/0/L5Sc+/f+/n8+yqutzh7xhJdz3yZUcTDhcYMmSIWy0H7i277ufs+r7vV0Zvt/Ren6uXz+cKqZFdPy+Z7X27S738/Zsxy3OX/Z2ROAUxi4iKilLevHl15coVt0j/QGbG9yl7YX+nL7avM7aH+8oM+8bdanSHetyhhtTiCFgW4ePjo5CQEPn4+FhdCpDp8X3KXtjf6Yvt64zt4b4yw75xtxrdoR53qCG1OAIGAAAAABmEI2AAAAAAkEEIYAAAAACQQQhgAAAAAJBBCGAAAAAAkEEIYAAAAACQQQhg2UCHDh2UP39+PfPMM1aXAgAAAKS7yMhI1axZU4GBgapSpYo+/PBDq0tyoA19NrB161ZdvXpVixcv1ooVK6wuBwAAAEhX8fHxunnzpvz8/BQdHa0qVapo165dKliwoNWlcQQsO2jUqJHy5MljdRkAAABAhvD09JSfn58k6ebNmzIMQ+5y3IkA5uZ++OEHtW3bVsWLF5fNZtPq1asTjZk3b57KlCkjX19f1alTR7/++mvGFwoAAAC4iCt+B46MjFS1atVUsmRJvfrqqypUqFAGVX93BDA3Fx0drWrVqmnevHlJPr9s2TIFBwcrJCREe/bsUbVq1dSyZUudP38+gysFAAAAXMMVvwPny5dP+/fv1/Hjx7VkyRKdO3cuo8q/K64By0RsNptWrVql9u3bO+bVqVNHtWrVUmhoqCQpISFBpUqV0tChQzV69GjHuK1btyo0NJRrwAAAAJCp3M/vwHYvvviimjRp4hZN6TgClonFxMRo9+7datasmWOeh4eHmjVrph07dlhYGQAAAJA+UvI78Llz53T16lVJ0pUrV/TDDz+oYsWKltR7Jy+rC0DaXbx4UfHx8SpatKjT/KJFi+rw4cOOn5s1a6b9+/crOjpaJUuW1PLly1W3bt2MLhcAAAC4byn5HfjkyZMaOHCgo/nG0KFDVbVqVSvKTYQAlg1s2rTJ6hIAAACADFO7dm3t27fP6jKSxCmImVihQoXk6emZ6ILCc+fOKSAgwKKqAAAAgPST2X8HJoBlYjly5FCNGjW0efNmx7yEhARt3ryZUwwBAACQJWX234E5BdHNXbt2TX/88Yfj5+PHj2vfvn0qUKCASpcureDgYPXu3Vs1a9ZU7dq1NWvWLEVHR6tv374WVg0AAACkXVb+HZg29G5u69ataty4caL5vXv31qJFiyRJoaGheueddxQREaHAwEDNmTNHderUyeBKAQAAANfIyr8DE8AAAAAAIINwDRgAAAAAZBACGAAAAABkEAIYAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYAAAAAGQQAhgAAAAAZBACGAAAAABkEAIYACDT279/v0JDQ60uAwCAeyKAAQAytRs3bqh79+6qWLFiql9rGIaGDx+uQYMGpUNlAAAkRgADAGQ6586dU5kyZfSvf/1Lu3fv1htvvKHmzZsnOfbEiROy2Wzat2+f0/wFCxaoePHiypEjhxo0aKATJ06kf+EAgGyPAAYAcBsXLlzQ4MGDVbp0afn4+CggIEAtW7bUTz/95DRu0KBBmjFjhkaNGqWPPvpIHTt2TNV6jh49qoULF+rQoUPat2+fihcvrjJlytxX7RMmTFBgYOB9LcOV7MGzSJEiunr1qtNzgYGBmjBhgjWFAUA252V1AQAA2HXq1EkxMTFavHixypUrp3Pnzmnz5s26dOmSY4xhGFqwYIEKFy4sSWrbtm2q11OhQgX9/PPPkqSNGze6png3dfXqVU2fPl0TJ060uhQAgDgCBgBwE5GRkdq2bZumTp2qxo0b64EHHlDt2rU1ZswYPf30045xp0+f1vPPP6/cuXPL399fQUFBOnfuXIrXEx8fr/79+6ts2bLKmTOnKlasqNmzZzuN6dOnj9q3b6/p06erWLFiKliwoIYMGaLY2Ngkl7lo0SJNnDhR+/fvl81mk81m06JFi5I8/TEyMlI2m01bt26VJG3dulU2m02bN29WzZo15efnp8cff1xHjhxxWsebb76pIkWKKE+ePHr++ec1evToFB1xGzp0qGbOnKnz58+neBsBANIPAQwA4BZy586t3Llza/Xq1bp582aSYxISEtSuXTtdvnxZ33//vcLCwvTnn3+qS5cuKV5PQkKCSpYsqeXLl+vgwYMaP368XnvtNX3xxRdO47Zs2aJjx45py5YtWrx4sRYtWqRFixYlucwuXbro5Zdf1sMPP6zw8HCFh4enqiZJev311zVjxgzt2rVLXl5e6tevn+O5zz77TG+99ZamTp2q3bt3q3Tp0nr//fdTtNxu3brpwQcf1KRJk1JVDwAgfXAKIgDALXh5eWnRokUaMGCA5s+fr+rVq6thw4bq2rWrHnnkEUnS5s2b9dtvv+n48eMqVaqUJOmTTz7Rww8/rJ07d6pWrVr3XI+3t7fT6Xhly5bVjh079MUXX+jZZ591zM+fP79CQ0Pl6empSpUqqU2bNtq8ebMGDBiQaJk5c+ZU7ty55eXlpYCAgDS9/7feeksNGzaUJI0ePVpt2rTRjRs35Ovrq7lz56p///7q27evJGn8+PHauHGjrl27ds/l2mw2TZkyRW3bttWIESNUvnz5NNUHAHANjoABANxGp06ddPbsWX399ddq1aqVtm7dqurVqzuOPB06dEilSpVyhC9Jqly5svLly6dDhw6leD3z5s1TjRo1VLhwYeXOnVsLFizQqVOnnMY8/PDD8vT0dPxcrFixdD2Nzx4y7euS5FjfkSNHVLt2bafxd/58Ny1btlS9evU0btw4F1QKALgfBDAAgFvx9fVV8+bNNW7cOG3fvl19+vRRSEiIy5a/dOlSvfLKK+rfv782btyoffv2qW/fvoqJiXEa5+3t7fSzzWZTQkJCqtbl4WH+M2sYhmNecteR3b4+m80mSale391MmTJFy5Yt0969e122TABA6hHAAABurXLlyoqOjpYkPfTQQzp9+rROnz7teP7gwYOKjIxU5cqVU7S8n376SY8//rhefPFFPfroo3rwwQd17Nix+64zR44cio+Pd5pn79QYHh7umHfn/chSomLFitq5c6fTvDt/vpfatWurY8eOGj16dKrXDwBwHa4BAwC4hUuXLqlz587q16+fHnnkEeXJk0e7du3StGnT1K5dO0lSs2bNVLVqVT333HOaNWuW4uLi9OKLL6phw4aqWbNmitZToUIFffLJJ9qwYYPKli2rTz/9VDt37lTZsmXvq/4yZcro+PHj2rdvn0qWLKk8efIoZ86ceuyxxzRlyhSVLVtW58+f19ixY1O97KFDh2rAgAGqWbOmHn/8cS1btkz//e9/Va5cuVQt56233tLDDz8sLy/++QcAq3AEDADgFnLnzq06dero3XffVYMGDVSlShWNGzdOAwYMUGhoqCTz1LyvvvpK+fPnV4MGDdSsWTOVK1dOy5YtS/F6XnjhBXXs2FFdunRRnTp1dOnSJb344ov3XX+nTp3UqlUrNW7cWIULF9bnn38uSVq4cKHi4uJUo0YNDR8+XG+++Waql/3cc89pzJgxeuWVV1S9enUdP35cffr0ka+vb6qW869//Uv9+vXTjRs3Ul0DAMA1bMbtJ6YDAIBMoXnz5goICNCnn35qdSkAgFTgHAQAANzc9evXNX/+fLVs2VKenp76/PPPtWnTJoWFhVldGgAglTgCBgCAm/vnn3/Utm1b7d27Vzdu3FDFihU1duxYdezY0erSAACpRAADAAAAgAxCEw4AAAAAyCAEMAAAAADIIAQwAAAAAMggBDAAAAAAyCAEMAAAAADIIAQwAAAAAMggBDAAAAAAyCAEMAAAAADIIP8H9UZwfXO0a6EAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "N = 10, Phương sai của ước lượng X: 0.0249\n",
            "N = 40, Phương sai của ước lượng X: 0.0041\n",
            "N = 90, Phương sai của ước lượng X: 0.0031\n",
            "N = 160, Phương sai của ước lượng X: 0.0016\n",
            "N = 250, Phương sai của ước lượng X: 0.0010\n",
            "N = 490, Phương sai của ước lượng X: 0.0005\n",
            "N = 640, Phương sai của ước lượng X: 0.0003\n",
            "N = 810, Phương sai của ước lượng X: 0.0003\n",
            "N = 1000, Phương sai của ước lượng X: 0.0002\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Số lần tung đồng xu\n",
        "N_values = [10, 40, 90, 160, 250, 490, 640, 810, 1000]\n",
        "\n",
        "# Danh sách để lưu phương sai\n",
        "variances = []\n",
        "\n",
        "# Mô phỏng cho mỗi giá trị N\n",
        "for N in N_values:\n",
        "    estimates = []\n",
        "\n",
        "    # Chạy mô phỏng 100 lần\n",
        "    for _ in range(100):\n",
        "        # Tung đồng xu N lần (0 là mặt Tail, 1 là mặt Head)\n",
        "        tosses = np.random.binomial(1, 0.5, N)\n",
        "        # Tính giá trị ước lượng cho X\n",
        "        estimate = np.mean(tosses)\n",
        "        estimates.append(estimate)\n",
        "\n",
        "    # Tính phương sai của các ước lượng\n",
        "    variance = np.var(estimates, ddof=1)\n",
        "    variances.append(variance)\n",
        "\n",
        "# Vẽ biểu đồ phương sai so với N\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.plot(N_values, variances, marker='o', linestyle='-', color='b')\n",
        "plt.title('Phương sai của ước lượng X (tung đồng xu) theo N')\n",
        "plt.xlabel('Số lần tung N')\n",
        "plt.ylabel('Phương sai của ước lượng')\n",
        "plt.xscale('log')  # Sử dụng thang đo logarit cho trục x\n",
        "plt.grid()\n",
        "plt.xticks(N_values)\n",
        "plt.show()\n",
        "\n",
        "# In kết quả phương sai\n",
        "for N, variance in zip(N_values, variances):\n",
        "    print(f\"N = {N}, Phương sai của ước lượng X: {variance:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5aj48-pJEGFp"
      },
      "source": [
        "---"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.11.1"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}