Progress/Week_7/Linear_Regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Linear_Regression.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fss48S45kFAs"
      },
      "source": [
        "<h1 align='center'>Mô Hình Linear Regression Đơn Và Đa Biến</h1>\n",
        "<h1 align='center'>Đỗ Trọng Khánh - 19521676</h1>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hMbWOrJMkg6C"
      },
      "source": [
        "# Connect Driver"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "41COjIrPkae6",
        "outputId": "c5c24b0b-d4dd-416b-af14-c4a92538bcbc"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/gdrive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /gdrive; to attempt to forcibly remount, call drive.mount(\"/gdrive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nsFZ9Ib3kvnN"
      },
      "source": [
        "# Import Libary"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FLfDiaHckwpn"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from pandas import DataFrame\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.metrics import accuracy_score, classification_report\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.preprocessing import OneHotEncoder, LabelEncoder\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.metrics import r2_score\n",
        "import seaborn as sns\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "O2xHn_phOBZC"
      },
      "source": [
        "# Simple linear regression"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GkUmpe2XPPT1"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DUrQ1s3fPX40"
      },
      "source": [
        "data_link = '/gdrive/MyDrive/Colab Notebooks/Lập trình Python cho Máy học/Week 7/Data/Salary_Data.csv'\n",
        "data = pd.read_csv(data_link)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SHOtVB-oQAOv"
      },
      "source": [
        "df = pd.DataFrame(data)\n",
        "X_data = df.iloc[:, 0:-1];\n",
        "Y_data = df.iloc[:, -1:];\n",
        "\n",
        "X_data_np = X_data.to_numpy()\n",
        "Y_data_np = Y_data.to_numpy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GDY74LTqu63B"
      },
      "source": [
        "## Visualize Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W-yYIPxjvT8I"
      },
      "source": [
        "def visualize_data(X_data, Y_data, model):\n",
        "    sns.set()\n",
        "    a = model.coef_[0] # Hệ số dự đoán cho từng dữ liệu\n",
        "    b = model.intercept_[0] # Lấy hằng số của hàm quyết định\n",
        "    x_max = np.max(X_data)\n",
        "    x_min = np.min(X_data)\n",
        "\n",
        "    x = [i for i in np.arange(x_min, x_max, 0.01)]\n",
        "    y = a*x + b\n",
        "    plt.plot(x, y, color=\"red\")\n",
        "    plt.scatter(X_data, Y_data)\n",
        "    plt.xlabel('YearsExperience')\n",
        "    plt.ylabel('Salary')\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gStMo8GGRHTE"
      },
      "source": [
        "## Tranning and evaluate model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WDWFBuFDRJLL"
      },
      "source": [
        "### Simple Linear Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B7FSLymbRR8Y"
      },
      "source": [
        "from sklearn.linear_model import LinearRegression\n",
        "LinearRegression_model = LinearRegression()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        },
        "id": "DOr6TMw0Rlcx",
        "outputId": "5e0ab0a7-8aba-47c2-b115-8b829cbdf801"
      },
      "source": [
        "kfold = KFold(5, True, 1);\n",
        "score_list = [];\n",
        "\n",
        "# Sử dụng kfold để chia dữ liệu và đánh giá mô hình\n",
        "for train, test in kfold.split(X_data_np, Y_data_np):\n",
        "    X_train, X_test = X_data_np[train], X_data_np[test];\n",
        "    Y_train, Y_test = Y_data_np[train], Y_data_np[test];\n",
        "    \n",
        "    LinearRegression_model.fit(X_train, Y_train);\n",
        "    y_pred = LinearRegression_model.predict(X_test)\n",
        "    score_list.append(r2_score(y_pred, Y_test, sample_weight=None))\n",
        "\n",
        "print('\\tR2_Score:', np.mean(score_list))\n",
        "visualize_data(X_train, Y_train, LinearRegression_model)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\tR2_Score: 0.8742256057843416\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "L3S3VfMCNcpP"
      },
      "source": [
        "# Multiple linear regression"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gdOQVx8yk0hf"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iXgV38SSky0A"
      },
      "source": [
        "data_link = '/gdrive/MyDrive/Colab Notebooks/Lập trình Python cho Máy học/Week 7/Data/50_Startups.csv'\n",
        "data = pd.read_csv(data_link)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "RDah3t21lA1X",
        "outputId": "2ad805bb-9a2f-4f5f-d155-ae87f139b1aa"
      },
      "source": [
        "data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>R&amp;D Spend</th>\n",
              "      <th>Administration</th>\n",
              "      <th>Marketing Spend</th>\n",
              "      <th>State</th>\n",
              "      <th>Profit</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>165349.20</td>\n",
              "      <td>136897.80</td>\n",
              "      <td>471784.10</td>\n",
              "      <td>New York</td>\n",
              "      <td>192261.83</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>162597.70</td>\n",
              "      <td>151377.59</td>\n",
              "      <td>443898.53</td>\n",
              "      <td>California</td>\n",
              "      <td>191792.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>153441.51</td>\n",
              "      <td>101145.55</td>\n",
              "      <td>407934.54</td>\n",
              "      <td>Florida</td>\n",
              "      <td>191050.39</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>144372.41</td>\n",
              "      <td>118671.85</td>\n",
              "      <td>383199.62</td>\n",
              "      <td>New York</td>\n",
              "      <td>182901.99</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>142107.34</td>\n",
              "      <td>91391.77</td>\n",
              "      <td>366168.42</td>\n",
              "      <td>Florida</td>\n",
              "      <td>166187.94</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   R&D Spend  Administration  Marketing Spend       State     Profit\n",
              "0  165349.20       136897.80        471784.10    New York  192261.83\n",
              "1  162597.70       151377.59        443898.53  California  191792.06\n",
              "2  153441.51       101145.55        407934.54     Florida  191050.39\n",
              "3  144372.41       118671.85        383199.62    New York  182901.99\n",
              "4  142107.34        91391.77        366168.42     Florida  166187.94"
            ]
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bJx_9LASlCXy"
      },
      "source": [
        "df = pd.DataFrame(data)\n",
        "X_data = df.iloc[:, 0:-2];\n",
        "Y_data = df.iloc[:, -1:];"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oq7ij-uMlabd"
      },
      "source": [
        "x_profit_state = df['State'].to_numpy().reshape(-1, 1)\n",
        "\n",
        "onehot_encoder = OneHotEncoder()\n",
        "x_profit_state = onehot_encoder.fit_transform(x_profit_state).toarray()\n",
        "\n",
        "X_data = np.hstack([X_data, x_profit_state])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "45-vgQCDoAI1"
      },
      "source": [
        "Y_data = Y_data.to_numpy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Iw7dspx_om6l"
      },
      "source": [
        "## Tranning and evaluate model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LAMOiAg3opoJ"
      },
      "source": [
        "### Multiple Linear Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Agq_h4_poPaX"
      },
      "source": [
        "from sklearn.linear_model import LinearRegression\n",
        "LinearRegression_model = LinearRegression()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 304
        },
        "id": "kBR-r40foRBj",
        "outputId": "65243be0-299b-44eb-f595-4c22ce47b063"
      },
      "source": [
        "kfold = KFold(5, True, 1);\n",
        "\n",
        "score_list = [];\n",
        "for train, test in kfold.split(X_data, Y_data):\n",
        "    X_train, X_test = X_data[train], X_data[test];\n",
        "    Y_train, Y_test = Y_data[train], Y_data[test];\n",
        "    \n",
        "    LinearRegression_model.fit(X_train, Y_train);\n",
        "    y_pred = LinearRegression_model.predict(X_test)\n",
        "    score_list.append(r2_score(y_pred, Y_test, sample_weight=None))\n",
        "\n",
        "print('\\tR2_Score:', np.mean(score_list))\n",
        "\n",
        "# Visualising data\n",
        "sns.set()\n",
        "x = np.arange(0, 175000).reshape((-1, 1))\n",
        "plt.xlabel('R&D Spend')\n",
        "plt.ylabel('Profit')\n",
        "plt.scatter(X_train[:, 0], Y_train)\n",
        "a = LinearRegression_model.coef_[0][0]\n",
        "b = LinearRegression_model.intercept_\n",
        "y = a*x +b\n",
        "plt.plot(x, y, color=\"red\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\tR2_Score: 0.8898522301315502\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAacAAAEOCAYAAAAt/Ni+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deVxU9f7H8dcMMGyKA4gKmFmURppKUmpqGC1a16U0y90yu91KM71alKk3XAG3NIzylmlR3swtc+9nmpX7knuZmangxr7DzJzfH1zmirLDzJwDn+fj0ePRnO85c94zHPlwzvme71enKIqCEEIIoSJ6RwcQQgghbiTFSQghhOpIcRJCCKE6UpyEEEKojhQnIYQQqiPFSQghhOrYpTilpKTw4osv0r17d3r16sWoUaNITk4G4PDhw/Tu3Zvu3bszYsQIkpKSrNvZu00IIYRKKHaQkpKi7N692/p61qxZyltvvaWYzWblkUceUfbt26coiqLExsYqERERiqIodm8TQgihHnY5czIajXTo0MH6ul27diQkJHDs2DFcXV0JDQ0FYMCAAWzatAnA7m1CCCHUw9neO7RYLHz55ZeEh4eTmJhIQECAtc3HxweLxUJqaqrd24xGY4U/Q0pKFhZL5QfW8PWtR1JSZqW3cwQtZQVt5dVSVtBWXi1lBW3lrU5WvV6Ht7dnpbaxe3GaOnUqHh4eDBkyhK1bt9p79zWisl/y9Xx969VgEtvSUlbQVl4tZQVt5dVSVtBWXntmtWtxioqK4ty5c8TFxaHX6/H39ychIcHanpycjF6vx2g02r2tMpKSMqt05uTnV5+rVzMqvZ0jaCkraCuvlrKCtvJqKStoK291sur1ukoXNrt1JZ87dy7Hjh0jNjYWg8EAQOvWrcnNzWX//v0ALF++nB49ejikTQghhHroFMX2o5KfPn2anj170rx5c9zc3ABo2rQpsbGxHDx4kClTppCXl0dgYCAxMTE0bNgQwO5tFSVnTuqjpbxaygrayqulrKCtvPY+c7JLcaptpDipj5byaikraCuvlrKCtvLW2st6QgghREXZvbeeEEII29h1/BKrdpwhKT0PXy9X+oYF0alVE0fHqhIpTkIIoRLVKS67jl9i6cZT5JssACSl57F04ymAahUop+PH8Jw1Fa5dgQ3bQKer8ntVhlzWE0IIFSgqLknpecD/isuu45cqtP2qHWeshalIvsnCqh1nqpTH6Y/fqf+PEfg89AAuu3fBiBF2K0wgxUkIIVShusWlqKhVdHlp9BcvUO+fr+Hd+T5cN20g6/XxJO/7BV55pVLvU11yWU8IIVSgusXF18u1xHV9vVwrtL3u2jU83puD+6f/BkUhZ8SLZI8Zj9KoUYW2r2ly5iSEECpQWhGpaHHpGxaEwbn4r3SDs56+YUFlbqdLT8Mjajo+97XBffEH5PbtT/Kug2RNj3ZYYQI5cxJCCFXoGxZUrEMDVKy4FCnq9FDhDhXZ2bh/shiPhXPRp6SQ2/spst+ciPnOFtX+LDVBipMQQqhApYtLKe9R7vr5+bjFL8NjbjROly+RH/4IWW9PxtSmXXXi1zgpTkIIoRIVKi5VZTbjumoFntEzcDr3JwUdOpHx0RIKOnW2zf6qSYqTEELUZoqCYeN6PGdNxfnUSQpatyHzy6/JD3/Url3DK0uKkxBC1BC1jdDg8sN2PGe8i8vBA5juuJO0fy8lv2cf0Ku/L5wUJyGEqAG2GqGhKpz378Vz5lQMO3dgDmxKxvxYcp8ZCM7a+ZWv/vIphBAaUNMjNFSF04njeA0bgPcTj+B88jiZ06NI3n2I3EFDNVWYQM6chBCiRtTUCA1VoT/7B57RM3BdtQKlvhdZb00i+8WXoZ52poC/kRQnIYSoAdUdoaEq9IkJeMyJxu2LZeDiQs7osWS/+hqKt4/N9mkvUpyEEKIGVPch2srQJSXhsWAu7ksWg9lM7vARZL8+HktjbU6PURIpTkIIUQNq4iHa8ugy0nGPi8X9g/fRZWeR138AWRPewtLs1hrbh1pIcRJCiBpis4doc3JwX/JvPBbMQZ+cTF7PPmS9ORFzy7tqfl8qYbfiFBUVxebNm7l48SLr1q2jRYsWXLhwgVdffdW6TkZGBpmZmezduxeA8PBwDAYDrq6F12zHjx9P165dATh8+DCTJ08mLy+PwMBAYmJi8PX1rVabEEKoSkEBbl9+jsecKJwSE8h/6GGy3pqEqd29jk5mc3brSv7www8THx9PYGCgdVnTpk1Zu3at9b+HH36Ynj17FttuwYIF1vaiwmSxWJgwYQKTJ09m8+bNhIaGMnv27Gq1CSGEalgsuK78Cp/OodQfPwZL01tIXbOBtP+srhOFCexYnEJDQ/H39y+1PT8/n3Xr1tGvX79y3+vYsWO4uroSGhoKwIABA9i0aVO12oQQwuEUBcOmDXg/1Bmvl0eieNYjLf4rUr/dQsEDXRydzq5Uc89p27ZtNG7cmFatWhVbPn78eBRFoX379owbNw4vLy8SExMJCAiwruPj44PFYiE1NbXKbUaj0fYfUgghSuHy4w94Tn8XlwP7MN0eRPpHS8jr/ZQmhhqyBdUUp5UrV9501hQfH4+/vz/5+flMnz6dyMhIVVyG8/Wt+oNtfn71azCJbWkpK2grr5aygrbyaikrgN/ZkzBxInz3HTRtCosX4zx8OF4uLo6OdhN7freqKE6XL19m3759REdHF1tedBnQYDAwaNAgXn75ZevyhIQE63rJycno9XqMRmOV2yojKSkTi0Wp9Of086vP1asZld7OEbSUFbSVV0tZQVt5tZTV6eQJfObNgjVrsPj6kj11JjnDXwA3N0jNBXIdHbGY6ny3er2u0n/Uq+J8cfXq1YSFheHt7W1dlp2dTUZG4RehKAobNmwgODgYgNatW5Obm8v+/fsBWL58OT169KhWmxBC2IP+z7PUf/XveHfrBNu2kfXmRJL3HSHnpVcLC5MA7HjmNG3aNLZs2cK1a9d4/vnnMRqNrF+/HigsThMnTiy2flJSEqNHj8ZsNmOxWAgKCmLKlCkA6PV6oqOjmTJlSrEu4dVpE0IIW9JfSsRjbjRuny8FZ2dyXh2Dx7/eIdticHQ0VdIpilL561N1nFzWUx8t5dVSVtBWXjVm1SUn4bFwPu4ffwgmE7lDnyN77AQsTfxVmbc09r6sp4p7TkIIUdvoMjNw/3AR7osWosvMIO/pZwuHGmp+m6OjaYIUJyGEqKISZ74NMuL+6b/xeG8O+qQk8p7oRVbEO5jvCnZ0XE2R4iSEEFVw48y3qSlZXJz1HvUOrsT96iXywx4i6+3JmELaOzipNklxEkKIKiia+VanWOjy608M/vlLAlMT+L3pXTRc9W8Kujzo6IiaJsVJCCGqICktl/vO7mfoj/Hcdu1Pzja8lcg+b7Pv9vv4RApTtUlxEkKISnL5+Ufmfj2RO8+fIKFBE2KeGMfOll1QdHqbznxbl0hxEkKICnI+fBDPGZEYtm/DtWFj4h57hU3B4ZidCn+V2mrm27pIipMQQpTD6ddTeM6ahuv6b7D4+JD57gxynnuBgD/SMNpw5tu6TIqTEEKUQn/uTzxjZuL69X9QPDzJmvAWOf94FaW+FwCdWrlLMbIRKU5CCHED/eVLeMyLwe2zT8HJiZx/jCJ79FgUmTXbbqQ4CSHEf+lSkvF4/z3c/x0HBQXkDh5O9rgJWPwDyt9Y1CgpTkIIkZmJx+IPcI9dgC4jnby+/QuHGrpdOjc4ihQnIUTdlZuL+7JP8Jg/G/21axxs2YlP+jxLZlBL+uZ40snR+eowKU5CiDqjaCy8lNRsev+xk4G7/4PHlUSuhnRkTo83Od7ozsIV0/NYuvEUgHR4cBApTkKIOmHX8Uss23CC+078yOCfv6BpSgK/+bfg2uyZfJTZmKT0vGLr55ssrNpxRoqTg0hxEkLUforC74u/JGrrpwRdPcs532ZM6/0We4LuxzfT7abCVKS05cL2pDgJIWo1l90/4zn9Xcbt2UVig8bMfnwsO1t2waJ3ArA+QFtSIZKhiBxHipMQolZyPnK4cKihbd9hbtyEpX8bxZo7wjA5uRRbr2hkh+unvwAZisjRpDgJIWoVp99+xTNqOq7r1mDx9iZzyjRyRryI9x9p6DeeghIKUNF9pZsmDpT7TQ5jt+IUFRXF5s2buXjxIuvWraNFixYAhIeHYzAYcHUtPH0eP348Xbt2BeDw4cNMnjyZvLw8AgMDiYmJwfe/T2jbok0IoV3683/BG7PxXrYMxd2DrPERhUMNeTUACocagtILUKdWTaQYqYhOURTFHjvav38/gYGBDB48mLi4uGLF6frXRSwWC927d2fmzJmEhoayaNEizp8/z8yZM23SVhlJSZlYLJX/2vz86nP1akalt3MELWUFbeXVUlZQf17d5ct4vDcb96WfoNPryX7+RbJfG4fSsKGjo5VL7d/t9aqTVa/X4etbr3LbVGlPVRAaGoq/v3+F1z927Biurq6EhoYCMGDAADZt2mSzNiGEtuhSU/Cc/i6+HdrivuTf5A4YAr//TlbkDE0UJlE2VdxzGj9+PIqi0L59e8aNG4eXlxeJiYkEBPxvPCsfHx8sFgupqak2aTMajRXOW9m/AK7n51e/ytvam5aygrbyaikrqCxvVhYsWADR0ZCWBgMHwr/+hfudhQ/Q+jk4XmWp6rsthz2zOrw4xcfH4+/vT35+PtOnTycyMpLZs2c7OlaZ5LKe+mgpr5aygory5uXh9tkSPOfNRn/1CnndHycrYhLmVq0L269mqCdrBWkpr70v6zm8OBVd6jMYDAwaNIiXX37ZujwhIcG6XnJyMnq9HqPRaJM2IUTpiob9cUhPNpMJ1xXL8YyZidOF8+R37krWp/GY7utgn/0Lh7DbPaeSZGdnk5FRWIkVRWHDhg0EBwcD0Lp1a3Jzc9m/fz8Ay5cvp0ePHjZrE0KUbPuB8yzdeMr6kGrSf8ed23X8km13bLFgWLcG77COeI15BYufH6kr1pK26lspTHWA3c6cpk2bxpYtW7h27RrPP/88RqORuLg4Ro8ejdlsxmKxEBQUxJQpUwDQ6/VER0czZcqUYt2+bdUmhCjZso0niz2cCjYed05RcPn+OzxnTMXlyGFMLe8ibUk8+U/0BJ2u5vcnVMluXclrE7nnpD5ayqulrAAvzNpGaUf7JxHhNbov59278JzxLobdP2Nu1pysN94ir98z4ORUoe219t1qKW+du+ckhFC3ht7uXE3JuWl5TY4753z0FzxmTsX1uy2YGzUmY9YccocMB4OhxvYhtMWh95yEEOo37PFgDM7Ff1XU1LhzTr+fpv6Lz+H9cFdc9u8lc1IkyXt/IXfEi1KY6jg5cxJClKlb+1tIz8it0d56+gvn8ZgThdvyeHB1I2vcG+S8PAqlgfScFYWkOAkhylVT487prl4tHGro048ByBn5Etmv/RPFT2uPzgpbk+IkhLA5XVoq7osW4PHhB5CXS+7AIWSPewNL01scHU2olBQnIYTtZGXh/vFHeLw/D31qKrlP9iX7zYmYg+50dDKhclKchBA1Lz8ft88+xWNeDE5XLpP3aPfCoYbuaePoZEIjpDgJIWqO2Vw41NDsWTj9dY78Tp1J//gzTB06OjqZ0BgpTkKI6lMUDOvX4TlrKs6//UpB2xAyYuZT0C1cRnUQVSLFSYhazqaDtioKLtu34TkjEpdfDmG6swVpH39Gfs/eUpREtUhxEqIW23X8Eks3nrKOjVc0aCtQ7QLlvHdP4VBDP/+I+ZZmpC/4gLz+Ayo81JAQZZHiJEQttmrHmRoftNXp2FE8Z0biunUzFr9GZMyMIXfIc+Bac8MZ3cihU3YIh5DiJEQtVjTNRUWXl8Xpj9/xiJqO2+qVWBoYyXznX+S88BJ4elY3ZplsefYn1EuKkxC1mK+Xa4mFqFKDtp4/T72Jk3H74jNwdSVr7HhyXnnNbkMN2eLsT6ifDPwqRC3WNyyoyoO26q5dw3PSW3Dnnbj95wtyRrxI0t4jZL812a5j4NXk2Z/QDjlzEqIWKzqzqMz9Gl16Gu6LFuL+4SJ0Odnw3HMkvzoOyy3N7BW7mBo5+xOaI8VJiFquwoO2ZmcXDjW0cG7hUEO9nyL7zYn4PNAeiwMnxOsbFlTsnhPU3JQdQr2kOAlR1+Xn4xa/DI+50ThdvkTew4+S/dYkTG3aOToZULWzP6F9ditOUVFRbN68mYsXL7Ju3TpatGhBSkoKb7zxBn/99RcGg4Fbb72VyMhIfHx8AGjZsiUtWrRAry+8Zh4dHU3Lli0B2LZtG9HR0ZjNZlq1asXMmTNxd3evVpsQdYrZjOvKr/CMnonTX39S0KETGYs/paDjA45OdpOamrJDaIfdOkQ8/PDDxMfHExgYaF2m0+kYOXIkmzdvZt26ddxyyy3Mnj272HbLly9n7dq1rF271lqYsrKymDRpEnFxcWzduhVPT08+/vjjarUJUWcoCmc//Iy8Nu3wGvUS5/Od2D1rManfbFJlYRJ1k92KU2hoKP7+/sWWGY1GOnToYH3drl07EhISyn2vH374gdatW9O8eXMABgwYwMaNG6vVJkRttOv4JSYs+okRs7YxYdFPnF62EuewLtw/6VUwmZjVcwKvDYwhJrkxu05cdnRcIaxUc8/JYrHw5ZdfEh4eXmz50KFDMZvNPPjgg4wePRqDwUBiYiIBAQHWdQICAkhMTASocpsQWlPeqAnXP7zaMuFXhq74nLbnj3LVy4/5j43m+7u7YdEXDjUkzw0JtVFNcZo6dSoeHh4MGTLEumz79u34+/uTmZnJhAkTiI2NZezYsQ5MWcjXt16Vt/Xzq1+DSWxLS1lBW3mrm3X7gfMs2/QreQVmoPCZn2WbfsWrvhvd2hfOLrvmx134J/7BkJ+/oOOZvaR4NODDh0ay6Z7umJxdbnrP5PS8UnPVpe/W3rSU155ZVVGcoqKiOHfuHHFxcdbOD4D1MmC9evXo378/S5YssS7fs2ePdb2EhATrulVtq4ykpEwsFqXS2/n51eeqA7vkVoaWsoK28tZE1k+/PW4tTEXyCsx8+u1xWjUzov/jDMPip/PgqZ1ku7qzrPNg1oX0JNdQeucfHy/XEnPVte/WnrSUtzpZ9Xpdpf+od/gIEXPnzuXYsWPExsZiMBisy9PS0sjNzQXAZDKxefNmgoODAejatStHjx7lzz//BAo7TTz++OPVahNCS0obHUG5mEC9f47Bp3Monc7s4ev7+zLyhY9Y0aG/tTDVc3eu8qgRQtiL3c6cpk2bxpYtW7h27RrPP/88RqOR+fPn8+GHH9K8eXMGDBgAQNOmTYmNjeWPP/5g8uTJ6HQ6TCYTISEhjBkzBig8k4qMjOSll17CYrEQHBzMxIkTq9UmhJbcOGqCV046T+9dSc/DG3DWQ+5zL/DDE8P5z56kmx5eHfhIC0CeGxLqplMUpfLXp+o4uaynPlrK6+dXn2+2n65WcSjq7OCUlcmTB9by5MFvcC3I42L3J/Gc9i6WZrda16tuEdLad6uVrKCtvPa+rKeKe05C1CXbD5yv9hQQnW5vQFD6TzRb+gH1c9LZd3cXMt54m3ue6ML143fLw6tCq6Q4CWFnyzaerPoUEAUFuH3xGR5zovC7lEj+Qw+T8vZkmrcNsWFiIexPipMQdnYtJafE5WVOAWE247r6azyjpuN07k8K7utARtzHFDzQxUYphXAsKU5C2FlDb3eullCgSpwCQlEwbN6I58xInE+ewNTqHtK+WEH+w4+BTmeHtEI4hsO7kgtR1wx7PLhCXblddu7A+MTDNBg2APLySP9oCSn/t5P8R7pLYRK1npw5CWFn3drfQnpGbqm96JwP7MNzxlQMO7djDggkY9775D47CJzln6uoO+RoF8IBSupF53TyBJ6zpuG68VssDRuSOXUmOcNfADc3B6UUwnEqfFnv/vvvL3F5p06daiyMEHXRL1v2cvqRJzGGdcK8bRsnR7xO8t5fyHnpVSlMos6q8JlTQUFBicssFksJawshrnf9w7B+3u482eU2OvsqZLzzL7p9+xVmvROr7nuKlaFPkV+/AcPPZdGplXYGBBWippVbnAYNGoROpyM/P5/BgwcXa7t06RIhIfJ8hRBluX7qCoDchMtY3ozDeHgDDUwmNt/zGP/p0J+UeoUzQCPTVwhRfnHq378/iqJw9OhRnn76aetynU6Hr68vHTt2tGlAIdSoMsMCrdpxhnyTBff8HPoc+IanDqzBLT+Xnfc8xGf3PcNl483blfnMkxB1QLnF6amnngKgbdu2BAXJqMVC3HgmVN7wQ+nJGfT5ZRP9935Ng5x0fr6jI/EPDOKvhs1K3UeJzzwJUYeUWZzWrFnDk08+CcChQ4c4dOhQietdf0YlRG1z41lSbr6pYsMPFRTgtjyexZ9G4pt+jUO3tuWzzkM43eTOMvcn01cIUU5xWr9+vbU4rV27tsR1dDqdFCdRa5V0llQaa5vFguualXhETcf57B8ktwphSuvXORjYukL7HP74XXK/SdR5ZRanojmWAD755BNcXG6e2lmI2qzoflFF+NY3YNiyEc8ZU3E+cYy/Gt/Gp09O5I+2nWlzR0M8T14mK9dc9nt4uUphEoJyitOECRM4ePAgAB06dLD+vxB1RUU7JoQkHOf1o1/T4PghMgNvZXHPf/L9nZ1RdHrIyOeno5dwcS57yCG5nCfE/5RZnPz8/Pj8888JCgrCbDaze/duSpqbUB7EFbXVjTPOXk+vg6DE04zY/QWt/ziE2T+AjDkLeD31Nq5kFT9DyjdZyDeVvR+ZjVaI/ymzOM2cOZMFCxawbNkyCgoKePvtt29aR6fT8X//9382CyiEI/UNCyp2z6nILUnnGfJTPA/8vpu8Bt5kRs4g57mR4ObGlVnbKrUPXy9XYl7pXJOxhdC8MovTvffey6effgrAo48+ytatW+2RSQjVKDqTKeqt1zjtMgN3LeehE9vJNbgR32kgO8P6MfUfj1m3Ke1sq567M/kFlmKFTi7lCVGyCg9fVFSYEhISuHz5Mk2aNMHf379C20ZFRbF582YuXrzIunXraNGiBQBnz54lIiKC1NRUjEYjUVFRNG/e3CFtQpSmU6smdG4Iu4eM4bGjW7Ho9KwO7cPK+/qS4e4F+cXXL+lsy+CsZ+Ajhcf9mh/PcjUlRy7lCVEGnVLSTaQSXL16lbFjx3L48GGMRiOpqam0bduWuXPn0rhx4zK33b9/P4GBgQwePJi4uDhrcRo2bBj9+vWjT58+rF27lpUrV7Js2TKHtFVGUlImFkuFvrZi/Pzqc/VqRqW3cwQtZYWazXv9c03NnPN5/a+t3Lb6MygoYEvrR/lPx2dILhpqiJIvy5U1gkRd/m5tTUtZQVt5q5NVr9fh61uvUttUuDi98sorBAQEMG7cODw8PMjOzmbu3LlcuHCBuLi4Cu0sPDzcWpySkpLo3r07e/bswcnJCbPZTIcOHdiyZQuKoti1zcfHp/zw15HipD41lbfouSZ9dha9D66j7/41uOfn8FPrMJbd/yyXjDdfLXix192VOvupq9+tPWgpK2grr72LU4Uv6x04cID33nvP+qyTh4cHb7zxBl27dq1cyv9KTEykcePGODk5AeDk5ESjRo1ITExEURS7tlW2OInaa913J+m+9xue2fM1xpw0dgV14PPOg/ir4a2lbiOX5YSoeRUuTg0aNODMmTPcdddd1mV//PEHXl5eNgmmZpX9C+B6fn7amQZBTVm3HzjPso0nuZaSQ0Nvd4Y9Hky39rcUW6e0vBXZFpOJU+/OY8aCaPwyrnG4WRs+6zyE3/xblJnLz9u9St+Tmr7bitBSXi1lBW3ltWfWChenkSNH8txzz/H0008TEBBAQkICq1atYsyYMVXasb+/P5cvX8ZsNlsvs125cgV/f38URbFrW2XJZT37+mzzKb4/lGB9fTUlh4VfHSY9I7fc+zg3Dj9007YWC67frEY/NZK7zp/lVJMWzO8+miPN2paby+Cs58kut1X6e1LTd1sRWsqrpaygrbz2vqxX4Zlwn3nmGebNm0dKSgrff/89KSkpzJkzh2effbbSQQF8fX0JDg7m22+/BeDbb78lODgYHx8fu7cJ9dp1/FKxwlSkaKDV8pQ0/FC+ycKq7b9j+G4zxkcexOvvz5OSpzC1z9tMGBhVZmHS/3eQB18vVxkDTwgbqlCHCLPZTPfu3dmwYQMGg6HSO5k2bRpbtmzh2rVreHt7YzQaWb9+PWfOnCEiIoL09HS8vLyIiori9ttvB7B7W2XImZP9TFj0U5lDCH0SEQ6UnndECQ/EtrpwnGE/fs7dCScx39qcrDcnMvQ3Hyx6pwplKtpnVanlu60oLeXVUlbQVl5VdohwcnLCycmJvLy8KhWnd955h3feeeem5UFBQaxYsaLEbezdJtSprMJUkTmPrn8gNujy7wz9MZ725w6RUt+XjJj55A4aCi4ueJdTBCuzTyFE9VX4st6wYcN4/fXX2bt3L3/99Rfnz5+3/ieErZRVDCoyskLfsCBuS7tIxLoo5seP587Lv7O02/P8tOJ7coePgP/2Pu0bFoTBuex/DjKagxD2U+EOEVOnTgXgp59+KrZcp9Nx8uTJmk0lxH+VNrbdQyEB5d7v0f91jsc+nEWvr74k19mVLzo+yw/d+vNE9zZ0uGHbG4cp8vVypU2QL0fOJFVoKnYhRM0qtzjl5OTwwQcf0K1bN+6++25eeuklXF3l0kZdVNaoB7ZSUtEob7+6y5fxnB+D27IloNeT89KrZL82jkd9fXm0nH1J8RFCHcotTpGRkRw7doyuXbuyZcsW0tLSmDRpkj2yCRUpaUbYpRtPAbZ/CLWiRUOXmoJH7ALcF38AeXnkDh5O9rgJWAICbZpPCFHzyr3ntHPnTj7++GPeeOMNFi9ezPfff2+PXEJlSu2SXYHu3DaXmQkzZuAT2gb3BXPJ6/E3kn/aT+bs+VKYhNCocs+csrOzadSoEVD44GxmZqbNQwn1Ka0nW0VnirWJvDzcl32Cx7zZcO0qBT2eIOvNdzC3au24TEKIGlFucbpxBlyTyXTTjLgyE27tV9ocRQ7pWm0y4bpiOZ4xM3G6cJ78zl0xfLOW9LI/8K8AABoGSURBVDukKAlRW5RbnHx9fYvNgGs0Gou9lplw64bS5iiya9dqiwXDt2vxnDUN599PUxByLxnz3qfgwW74NfICjTzMKIQoX7nFadu2yk05LWqnqvSaqzGKgmHbVjxmTMXl6C+Y7gom7dMvyH/8b6DT2X7/Qgi7q/BzTkI4oqu1y+6f8ZgRiWH3z5ibNSf9/Q/J6/cMOFVsqCEhhDZJcRKq5Hz0FzxmROL6f1sxN25CRtRccgcPgyoMnyWE0B4pTkJVnH4/jcesabh9sxqLtzeZk6eSM+JF8PBwdDQhhB1JcRKqoD//Fx5zonBbHg9u7mSNe4OcV0ajeDVwdDQhhANIcapjHDEEUVn7H3hPA7puWob70k9ApyPnxZfJfm0cip+f3TIJIdRHilMd4sghiG7cv2duJo//FM+D09fhai4gd/BQsse9gSWwqc1zCCHUT4pTHVLWEET2KE6rdpxBl5PN04fW02/fKurlZbGjZVc2PDqM8ZOrNqOyEKJ2kuJUh1R1CKLrL8XpdWBRqPwlwfx8Ou5YxbN7VuCdncre20P5/IFBnG1U+ZmIhRC1nxSnOqQqQxDdeCmwaHb6Cl8SNJsLhxqaPYt//HWOo01bMaN3BKcC7qrQ/oUQdVOFZ8IV2lfSbK/lDUFU0qXAImWOSq4oGNatxTusI16vvYzF24dd0Z/wr4EzihUmmV1WCFESh585XbhwgVdffdX6OiMjg8zMTPbu3Ut4eDgGg8E6ueH48ePp2rUrAIcPH2by5Mnk5eURGBhITEwMvr6+1Wqr7aoyBFF5l/xualcUXLZvw3NGJC6/HMLUoiVpn3xO/t96cYdOx3AH9xYUQmiDw4tT06ZNWbt2rfX19OnTMZvN1tcLFiygRYsWxbaxWCxMmDCBmTNnEhoayqJFi5g9ezYzZ86scltdUdkhiEq7FHh9exHnPbvxnBmJ4ecfMd/SjPQFH5DXf0CxoYZktlkhREWo6rJefn4+69ato1+/fmWud+zYMVxdXQkNDQVgwIABbNq0qVptomQlXQosUnRJzunoEbwG98e712M4n/6NjJmzSf75AHkDBssYeEKIKnH4mdP1tm3bRuPGjWnVqpV12fjx41EUhfbt2zNu3Di8vLxITEwkICDAuo6Pjw8Wi4XU1NQqtxmNxgrn9PWtV+XP6OdXv8rb2pufX316d6uPV303lm08ydWUHPR6HRaLgp+3Oy/d7UqHuEnwn/+AtzfMmoV+1Cjqe3riiE+pte9WS7SUV0tZQVt57ZlVVcVp5cqVxc6a4uPj8ff3Jz8/n+nTpxMZGcns2bMdmLBQUlImFotS/oo38POrz1WVzjl048gNz/VsRatmhQW7VTMjUS/9b0JJ/cULhUMN/etzFFc3sseOJ+eV11AaGCHbAtn2/4xq/m5vpKWsoK28WsoK2spbnax6va7Sf9Sr5rLe5cuX2bdvH7169bIu8/f3B8BgMDBo0CAOHjxoXZ6QkGBdLzk5Gb1ej9ForHJbXVbUXbzo3lJSeh7vr/iFXccvFVtPd/UqnpMi8OnQDrevviTnhb+TtPcXst+aXFiYhBCihqimOK1evZqwsDC8vb0ByM7OJiOjsEorisKGDRsIDg4GoHXr1uTm5rJ//34Ali9fTo8eParVVpeV1F08r8Bs7SauS0/DY9Y0fO5vi/viOHKffpbk3YfImhaF0qiRIyILIWo51VzWW716NRMnTrS+TkpKYvTo0ZjNZiwWC0FBQUyZMgUAvV5PdHQ0U6ZMKdYlvDptdVlpvfEyk9JxXzgfj4Vz0aemktunL9lvTsR8x512TiiEqGt0iqJU/uZJHVfb7jlNWPRTsQLlbC7gsaNbGbh3BcbMFPIeeYzstyZhuqetA1OWTa3fbUm0lBW0lVdLWUFbee19z0k1Z07CcfqGBbF04ylM+QWEnfqBQbuW0yTtMkltQkmZNh1Tx07lv4kQQtQgKU6CTnc3psnOLQQujCbw6l/82SSII1Nj8H+2DxadztHxhBB1kBSnOmzXsUR+XfI1fbZ8yv2XT5Nxy22kfbwMz7/1pnnjBpq53CCEqH2kONnR9gPn+fTb46oYV+7UVxu5PWoavc8f5Up9P+Y/Npqf2oQz9PZWdNKrphOnEKKOkuJkJ7uOX2LZpl/JKygcN9Des9AWcTp+DM9ZU+m6eSMpHg348KGRbLqnOyZnF7Bgt4kHhRCiLFKc7GTVjjPWwlTEnrPQ6v84g2f0DFxXf43i1YClXYawLqQneS5uxdYrbxRyIYSwB7l+YydVnYW2uvQJF6n3zzH4dA7FddN6cl4bR/K+X/j+kcE3FSaQif+EEOogZ052UpVZaKtDd+0aHgvm4r5kMVgs5Dw/kuwx41EaNwb+1338+pEhZOI/IYRaSHGyk75hQcXuOYFtioEuIx33RQtxj4tFl5NN3jMDyRofgaXZrcXWq8rEg0IIYS9SnOykU6smeNV3s11vvZwc3D9ZjMeCOehTUsjr9SRZb07E3KJlmZmkGAkh1EiKkx11a3+LdRqKGlNQgFv8MjzmRuN0KZH88EfIemsSprYhNbsfIYSwIylOWmU247r6azyjpuN07k8K7u9IxoefUNCps6OTCSFEtUlx0hpFwbBpA56zpuJ88gQFrduQ+cUK8h9+DGSoISFELSHFSUNcftiO54x3cTl4AFPQHaQv/pS8Xk+CjOgghKhlpDhpgPOBfXjOmIph53bMgU3JmB9L7jMDwVl+fEKI2kl+u6mY04njeM6ahuum9VgaNiRz2ixyho0At5sfnhVCiNpEipMK6c/+UTjU0KoVKPXqk/XWJLJffBnqVW6yLiGE0CopTiqiT0zAY24MbvFLwcWFnFGvkz1qDIq3j6OjCSGEXUlxUgFdUhIeC+fh/slHYDaTO+x5ssdOwNJYHpAVQtRNqihO4eHhGAwGXF0Lx5kbP348Xbt25fDhw0yePJm8vDwCAwOJiYnB19cXwCZt9qbLzMD9g/dx/+B9dNlZ5PUfUDjU0K3NHZJHCCHUQjV9kBcsWMDatWtZu3YtXbt2xWKxMGHCBCZPnszmzZsJDQ1l9uzZADZps4ftB84zYdFP/GPqRr4b8k/qh7TGM2YmBWEPkbJjNxkL46QwCSEEKipONzp27Biurq6EhoYCMGDAADZt2mSzNlvbdfwSHyw/QOiP3/DhklcYuGUxJ72b88MHX5O+5HPMLe+ySw4hhNACVVzWg8JLeYqi0L59e8aNG0diYiIBAQHWdh8fHywWC6mpqTZpMxorPuadr2/le839unQR89a8R0BqIif9WzLn8dc5dss9+F1z5xO/+pV+P3vwU2mu0mgpr5aygrbyaikraCuvPbOqojjFx8fj7+9Pfn4+06dPJzIykkcffdTRsUqVlJSJxaJUapv2ezaRY3An8smJ7Lst1DrU0NWUHK5ezbBFzGrx86uvylyl0VJeLWUFbeXVUlbQVt7qZNXrdZX+o14Vxcnf3x8Ag8HAoEGDePnllxk2bBgJCQnWdZKTk9Hr9RiNRvz9/Wu8zdaWPfumXScbFEIILXP4Pafs7GwyMgqrsaIobNiwgeDgYFq3bk1ubi779+8HYPny5fTo0QPAJm221jcsCFcXp2LLZOZZIYQomcPPnJKSkhg9ejRmsxmLxUJQUBBTpkxBr9cTHR3NlClTinX7BmzSZms2n2xQCCFqEZ2iKJW7eSIqfc9p1/FLrNpxhuT0PHw0UpS0dC0ctJVXS1lBW3m1lBW0lbdO3nOqzXYdv8TSjafIN1kASErPY+nGUwCqL1BCCOEoDr/nVNut2nHGWpiK5JssrNpxxkGJhBBC/aQ42VhJPfTKWi6EEEKKk82V1lVcupALIUTppDjZWN+wIAzOxb9m6UIuhBBlkw4RNlbU6UFrvfWEEMKRpDjZQadWTejUqommuo0KIYQjSXGyk13HL7Hmx11cTcmRB3CFEKIcUpzsQJ51EkKIypEOEXYgzzoJIUTlSHGyA3nWSQghKkeKkx3Is05CCFE5UpzsQJ51EkKIypEOEXZQ1OlhzY9npbeeEEJUgBQnO+nUqgm9u90pzzkJIUQFyGU9IYQQqiPFSQghhOpIcRJCCKE6Dr/nlJKSwhtvvMFff/2FwWDg1ltvJTIyEh8fH1q2bEmLFi3Q6wtraHR0NC1btgRg27ZtREdHYzabadWqFTNnzsTd3b1abUIIIdTB4WdOOp2OkSNHsnnzZtatW8ctt9zC7Nmzre3Lly9n7dq1rF271lqYsrKymDRpEnFxcWzduhVPT08+/vjjarUJIYRQD4cXJ6PRSIcOHayv27VrR0JCQpnb/PDDD7Ru3ZrmzZsDMGDAADZu3FitNiGEEOrh8Mt617NYLHz55ZeEh4dblw0dOhSz2cyDDz7I6NGjMRgMJCYmEhAQYF0nICCAxMREgCq3CSGEUA9VFaepU6fi4eHBkCFDANi+fTv+/v5kZmYyYcIEYmNjGTt2rINTgq9vvSpv6+dXvwaT2JaWsoK28mopK2grr5aygrby2jOraopTVFQU586dIy4uztoBwt/fH4B69erRv39/lixZYl2+Z88e67YJCQnWdavaVhlJSZlYLEqlt9PSZINaygrayqulrKCtvFrKCtrKW52ser2u0n/UO/yeE8DcuXM5duwYsbGxGAwGANLS0sjNzQXAZDKxefNmgoODAejatStHjx7lzz//BAo7TTz++OPVahNCCKEeDj9zOn36NB9++CHNmzdnwIABADRt2pSRI0cyefJkdDodJpOJkJAQxowZAxSeSUVGRvLSSy9hsVgIDg5m4sSJ1WoTQgihHjpFUSp/faqOk8t66qOlvFrKCtrKq6WsoK28dfKynhBCCHE9KU5CCCFUR4qTEEII1XF4h4i6YNfxS6zacYbk9Dx8ZKJBIYQolxQnG9t1/BJLN54i32QBICk9j6UbTwFIgRJCiFLIZT0bW7XjjLUwFck3WVi144yDEgkhhPpJcbKxpPS8Si0XQgghxcnmfL1cK7VcCCGEFCeb6xsWhMG5+NdscNbTNyzIQYmEEEL9pEOEjRV1epDeekIIUXFSnOygU6smdGrVRFNDlQghhCPJZT0hhBCqI8VJCCGE6khxEkIIoTpSnIQQQqiOdIioAr1e55Bt7U1LWUFbebWUFbSVV0tZQVt5q5q1KtvJZINCCCFURy7rCSGEUB0pTkIIIVRHipMQQgjVkeIkhBBCdaQ4CSGEUB0pTkIIIVRHipMQQgjVkeIkhBBCdaQ4CSGEUB0pTnZw9uxZnn32Wbp3786zzz7Ln3/+afN9pqSk8OKLL9K9e3d69erFqFGjSE5OBqBly5b06tWLPn360KdPH3799Vfrdtu2baNHjx48+uijvP766+Tk5FS7rSLCw8Pp0aOHNdPOnTsBOHz4ML1796Z79+6MGDGCpKQk6za2aKuICxcuWHP26dOH8PBw7r///jI/hz3zRkVFER4eTsuWLfntt9+sy8s6Du3dVl7eso5fcNwxXNp3a++fe0WPiZLylnX8OuKzlEoRNjd06FBlzZo1iqIoypo1a5ShQ4fafJ8pKSnK7t27ra9nzZqlvPXWW4qiKEqLFi2UzMzMm7bJzMxUHnjgAeXs2bOKoijK22+/rSxcuLBabRX10EMPKb/++muxZWazWXnkkUeUffv2KYqiKLGxsUpERITN2qpq2rRpyrvvvlvq57B33n379ikJCQk3ZSnrOLR3W3l5yzp+FcVxx3Bp3609f+6VOSZKy3u9649fe3+WskhxsrFr164p7du3V0wmk6IoimIymZT27dsrSUlJds2xadMmZfjw4YqilP4Pe8OGDcrf//536+sjR44oTzzxRLXaKqqkfxC//PKL8re//c36OikpSWnXrp3N2qoiLy9P6dChg3Ls2LFSP4ej8l6fpazj0N5tFcl7o+uPX0Vx/DFc0eKklmO4tHw3Hr/2/ixlkVHJbSwxMZHGjRvj5OQEgJOTE40aNSIxMREfHx+7ZLBYLHz55ZeEh4dblw0dOhSz2cyDDz7I6NGjMRgMJCYmEhAQYF0nICCAxMRE6+eoSltljB8/HkVRaN++PePGjbvpfX18fLBYLKSmptqkzWg0Vjrztm3baNy4Ma1atSr1c3h5eTk8b1nHoaIodm2r7HFf0vEL6juG7fVzr8ljuKTj156fpay8cs+pDpg6dSoeHh4MGTIEgO3bt7Nq1Sri4+P5/fffiY2NdXBCiI+P55tvvmHlypUoikJkZKSjI1XIypUr6devn/W1Vj+Hmt14/IL6jmGt/txvPH5BPZ9FipON+fv7c/nyZcxmMwBms5krV67g7+9vl/1HRUVx7tw55s+fj16vt2YCqFevHv379+fgwYPW5QkJCdZtExISrOtWta2iitY3GAwMGjSIgwcP3vS+ycnJ6PV6jEajTdoq6/Lly+zbt49evXqV+TmKljsyb1nHob3bKqOk47fo84B6jmF7/txr6pgo6fi192cpixQnG/P19SU4OJhvv/0WgG+//Zbg4GC7XNKbO3cux44dIzY2FoPBAEBaWhq5ubkAmEwmNm/eTHBwMABdu3bl6NGj1l5Vy5cv5/HHH69WW0VkZ2eTkZEBgKIobNiwgeDgYFq3bk1ubi779++3vm+PHj0AbNJWWatXryYsLAxvb+8yP4ca8pZ1HNq7raJKOn5BfcewvX/uNXVM3Hj8OuKzlEUmG7SDM2fOEBERQXp6Ol5eXkRFRXH77bfbdJ+nT5+mZ8+eNG/eHDc3NwCaNm3KyJEjmTx5MjqdDpPJREhICG+//Taenp4AfPfdd8TExGCxWAgODmbWrFl4eHhUq60858+fZ/To0ZjNZiwWC0FBQbzzzjs0atSIgwcPMmXKFPLy8ggMDCQmJoaGDRsC2KStMrp3787EiRN58MEHy/0c9sw7bdo0tmzZwrVr1/D29sZoNLJ+/foyj0N7t5WXd/78+SUev7GxsRw6dMhhx3BJWePi4uz+c6/oMVHasVDS8aumYxikOAkhhFAhuawnhBBCdaQ4CSGEUB0pTkIIIVRHipMQQgjVkeIkhBBCdaQ4CSFqxNChQ1mxYoWjY4haQoqTEDUgPDycNm3aEBISQufOnYmIiCArK6vYOj/++CM9evQgJCSE3r17c+LEiWLte/bs4a677iIkJISQkBAefPBBxowZw5EjR8rc94oVK6zv+8ADD/Diiy+SmZlZ459RCHuS4iREDYmLi+PQoUOsWbOGEydO8NFHHxVrj4iIYMSIERw8eJA5c+bg5eV103s0atSIQ4cOcfDgQb766ituv/12Bg8ezK5du0rc5969e5k3bx5z587l0KFDbNiwgSeeeMImn08Ie5JRyYWoYX5+fnTp0oWTJ08WW+7s7ExgYCA6nY4777yzzPfQ6XQ0adKEMWPGkJaWRkxMDKtWrbppvaNHj9KuXTvuvvtuAIxGI0899ZS1PSIiAoPBwPnz5zl8+DCtWrUiKiqKwMBAoHAUh2nTpnH8+HG8vb0ZM2aMtbhFRETg7u7OxYsX2bdvH3fccQdz5syhWbNmAPz0009MnTqVq1ev0qdPH+R5flGT5MxJiBp26dIldu7caf0lDoXjlLVp04Z33nmHCxcuVOr9Hn30UU6cOEF2dvZNbW3btuXHH39kwYIFHDhwgPz8/JvWWbduHa+88or1suH48eOBwnHURowYQc+ePfn555+ZN28e7777Lr///rt12w0bNjBq1Cj27dtHs2bNmDdvHlA4eOeoUaN4/fXX2b17N82aNbMOECpETZDiJEQNefXVVwkJCSEsLAwfHx9ee+01a9vixYvJyclh7NixPPfcc9YCtWLFCkaPHl3m+zZq1AhFUawDcl4vNDSUhQsXcuLECV566SU6dOjAzJkzraOBA3Tr1o377rsPg8HA2LFjOXz4MImJiWzfvp3AwED69euHs7Mzd999N927d2fTpk3WbR955BHatGmDs7MzvXv3tp4N/vDDD9x555306NEDFxcXhg8fXqXxCYUojVzWE6KGxMbG8sADD7B3717++c9/kpKSYr2vtGzZMhYuXEhISAhZWVkMGzaMZcuWcfDgQTp27Fjm+165cgWdTkf9+vVLbA8LCyMsLAyLxcKePXsYM2YMt912GwMGDACgSZMm1nU9PT1p0KABV65c4eLFixw5coTQ0FBru9lspnfv3tbX1xccNzc369nblStXir2vTqez2zQwom6Q4iREDbv//vvp27cvUVFRLFq0CCic2sFkMgEwcOBA0tLSGDZsGAATJ04s8/22bt3K3XffXe4o73q9nk6dOtGxY0dOnz5tXX7p0iXr/2dlZZGWlkajRo3w9/fnvvvuY8mSJZX+jH5+fsXeV1GUKs1+LERp5LKeEDYwfPhwfv75Z06dOgVAjx49iI6O5vz585hMJtq0aUNqaiouLi5YLJabtlcUhcuXL/P++++zYsUKxo0bV+J+vvvuO9avX09aWhqKonDkyBH27t1L27Ztrevs2LGD/fv3k5+fz3vvvUfbtm3x9/enW7du/Pnnn6xZs4aCggIKCgo4cuQIZ86cKffzhYWFcfr0abZs2YLJZGLZsmVcu3atit+WEDeT4iSEDfj4+NCnTx/r9OERERGEhoYyePBg7rvvPhYuXEhsbCx33XUXo0aNoqCgACi8XFb0nFO/fv347bff+Oyzz+jSpUuJ+2nQoAFfffUVjz32GPfeey8TJkzghRdeKHZprmfPnsTGxtKhQweOHz9OTEwMUDiL7Mcff8yGDRvo2rUrXbp0Yfbs2SV2qijp87333nvMmTOHDh06cO7cOe69997qfm1CWMl8TkLUYhERETRu3JixY8c6OooQlSJnTkIIIVRHipMQQgjVkct6QgghVEfOnIQQQqiOFCchhBCqI8VJCCGE6khxEkIIoTpSnIQQQqiOFCchhBCq8//dVriMMtLQdQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wXOVLuGJQiqc"
      },
      "source": [
        "# Linear Regression Position Salaries"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VFJL3G-grork"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HAZBZnNwrK49"
      },
      "source": [
        "data_link = '/gdrive/MyDrive/Colab Notebooks/Lập trình Python cho Máy học/Week 7/Data/Position_Salaries.csv'\n",
        "data = pd.read_csv(data_link)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IrVIGPpQsT0w",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "2d905078-370a-4daa-b58e-0a7d8ca1ade6"
      },
      "source": [
        "data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Position</th>\n",
              "      <th>Level</th>\n",
              "      <th>Salary</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Business Analyst</td>\n",
              "      <td>1</td>\n",
              "      <td>45000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Junior Consultant</td>\n",
              "      <td>2</td>\n",
              "      <td>50000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Senior Consultant</td>\n",
              "      <td>3</td>\n",
              "      <td>60000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Manager</td>\n",
              "      <td>4</td>\n",
              "      <td>80000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Country Manager</td>\n",
              "      <td>5</td>\n",
              "      <td>110000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            Position  Level  Salary\n",
              "0   Business Analyst      1   45000\n",
              "1  Junior Consultant      2   50000\n",
              "2  Senior Consultant      3   60000\n",
              "3            Manager      4   80000\n",
              "4    Country Manager      5  110000"
            ]
          },
          "metadata": {},
          "execution_count": 121
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oeOyVcmzs2dE"
      },
      "source": [
        "X_data = data.iloc[:, 1:2].values\n",
        "y_data = data.iloc[:, -1:].values"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ytHMdUS3vYqR"
      },
      "source": [
        "## Chuẩn hoá dữ liệu với StandardScaler"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t6Q4cAO9vef4"
      },
      "source": [
        "standard_ccaler = StandardScaler()\n",
        "sc_y_data = standard_ccaler.fit_transform(y_data)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0TsK4CK_tPOf"
      },
      "source": [
        "## Tranning and evaluate model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ermwTwzStSDu"
      },
      "source": [
        "### Model Support Vector Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bemrt09ctQ7S"
      },
      "source": [
        "from sklearn.svm import SVR\n",
        "SCR_model = SVR(kernel = 'poly')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        },
        "id": "OjaMw_tbtqEK",
        "outputId": "a16fa13d-2ba4-4d4b-d1ea-c93d9c269c5e"
      },
      "source": [
        "kfold = KFold(2, True, 1);\n",
        "score_list = [];\n",
        "\n",
        "# Sử dụng kfold để chia dữ liệu và đánh giá mô hình\n",
        "for train, test in kfold.split(X_data, sc_y_data):\n",
        "    X_train, X_test = X_data[train], X_data[test];\n",
        "    Y_train, Y_test = sc_y_data[train], sc_y_data[test];\n",
        "    \n",
        "    SCR_model.fit(X_train, Y_train);\n",
        "    y_pred = SCR_model.predict(X_test)\n",
        "    score_list.append(r2_score(Y_test, y_pred))\n",
        "\n",
        "print('\\tR2_Score:', np.mean(score_list))\n",
        "\n",
        "# Visualising data\n",
        "plt.scatter(X_data, sc_y_data, color='red')\n",
        "plt.plot(X_data,SCR_model.predict(X_data), color='blue')\n",
        "plt.title('Model Support Vector Regression')\n",
        "plt.xlabel('Position')\n",
        "plt.ylabel('Salary')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\tR2_Score: 0.5625498097233901\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UlGt77ck5SEE"
      },
      "source": [
        "### Model Random Forest Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WINJeRNF5RLT"
      },
      "source": [
        "from sklearn.ensemble import RandomForestRegressor\n",
        "RFR_model = RandomForestRegressor(n_estimators = 10, random_state = 0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        },
        "id": "IN0o1qSQ5x1G",
        "outputId": "d56b6f4b-c50c-4f54-ae23-517a333ac33a"
      },
      "source": [
        "kfold = KFold(2, True, 1);\n",
        "score_list = [];\n",
        "y_data = np.array(y_data).astype(int)\n",
        "# Sử dụng kfold để chia dữ liệu và đánh giá mô hình\n",
        "for train, test in kfold.split(X_data, y_data):\n",
        "    X_train, X_test = X_data[train], X_data[test];\n",
        "    Y_train, Y_test = y_data[train], y_data[test];\n",
        "    \n",
        "    RFR_model.fit(X_train, Y_train);\n",
        "    y_pred = RFR_model.predict(X_test)\n",
        "    score_list.append(r2_score(Y_test, y_pred))\n",
        "\n",
        "print('\\tR2_Score:', np.mean(score_list))\n",
        "\n",
        "# Visualising data\n",
        "x_grid = np.arange(min(X_data), max(X_data), 0.01)\n",
        "x_grid = x_grid.reshape((len(x_grid), 1))\n",
        "plt.scatter(X_data, y_data, color = 'red')\n",
        "plt.plot(x_grid, RFR_model.predict(x_grid), color = 'blue')\n",
        "plt.title('Model Random Forest Regression')\n",
        "plt.xlabel('Position')\n",
        "plt.ylabel('Salary')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\tR2_Score: 0.5083381253928466\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    }
  ]
}