MLP2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MLP2.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOn80IWaVnDVda0lAfg8hm9",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/nevermind78/Capstone/blob/main/MLP2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6UaS2mz8gPsU",
        "outputId": "2e949d6d-24ef-4d34-99b0-c2640acce21a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2021-12-09 19:40:01--  https://github.com/nevermind78/Capstone/blob/main/sepsis_Data2.csv\n",
            "Resolving github.com (github.com)... 140.82.114.3\n",
            "Connecting to github.com (github.com)|140.82.114.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: unspecified [text/html]\n",
            "Saving to: ‘sepsis_Data2.csv.1’\n",
            "\n",
            "sepsis_Data2.csv.1      [ <=>                ] 152.79K  --.-KB/s    in 0.08s   \n",
            "\n",
            "2021-12-09 19:40:02 (1.95 MB/s) - ‘sepsis_Data2.csv.1’ saved [156460]\n",
            "\n",
            "--2021-12-09 19:40:02--  https://github.com/nevermind78/Capstone/raw/main/finalized_model2.sav\n",
            "Resolving github.com (github.com)... 140.82.113.3\n",
            "Connecting to github.com (github.com)|140.82.113.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/nevermind78/Capstone/main/finalized_model2.sav [following]\n",
            "--2021-12-09 19:40:02--  https://raw.githubusercontent.com/nevermind78/Capstone/main/finalized_model2.sav\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 24109 (24K) [application/octet-stream]\n",
            "Saving to: ‘finalized_model2.sav.1’\n",
            "\n",
            "finalized_model2.sa 100%[===================>]  23.54K  --.-KB/s    in 0.001s  \n",
            "\n",
            "2021-12-09 19:40:02 (36.4 MB/s) - ‘finalized_model2.sav.1’ saved [24109/24109]\n",
            "\n"
          ]
        }
      ],
      "source": [
        "!wget https://github.com/nevermind78/Capstone/blob/main/sepsis_Data2.csv\n",
        "#!wget https://github.com/nevermind78/Capstone/raw/main/finalized_model.sav\n",
        "!wget https://github.com/nevermind78/Capstone/raw/main/finalized_model2.sav"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Code for MLP Classifier Algorithm"
      ],
      "metadata": {
        "id": "aG7WtX4XggP-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import pandas as pd\n",
        "from sklearn import preprocessing\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "Mwy0Y7qLghu1"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = pd.read_csv(\"https://raw.githubusercontent.com/nevermind78/Capstone/main/sepsis_Data2.csv\")\n",
        "dataset=dataset.drop(\"Unnamed: 0\",axis=1)\n",
        "dataset.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 226
        },
        "id": "4nk4NR8fgxFl",
        "outputId": "d4d11cd0-a9d9-4e38-84b0-a4a14a17ca32"
      },
      "execution_count": 37,
      "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>HR</th>\n",
              "      <th>O2Sat</th>\n",
              "      <th>Temp</th>\n",
              "      <th>SBP</th>\n",
              "      <th>MAP</th>\n",
              "      <th>DBP</th>\n",
              "      <th>Resp</th>\n",
              "      <th>EtCO2</th>\n",
              "      <th>BaseExcess</th>\n",
              "      <th>HCO3</th>\n",
              "      <th>FiO2</th>\n",
              "      <th>pH</th>\n",
              "      <th>PaCO2</th>\n",
              "      <th>SaO2</th>\n",
              "      <th>AST</th>\n",
              "      <th>BUN</th>\n",
              "      <th>Alkalinephos</th>\n",
              "      <th>Calcium</th>\n",
              "      <th>Chloride</th>\n",
              "      <th>Creatinine</th>\n",
              "      <th>Bilirubin_direct</th>\n",
              "      <th>Glucose</th>\n",
              "      <th>Lactate</th>\n",
              "      <th>Magnesium</th>\n",
              "      <th>Phosphate</th>\n",
              "      <th>Potassium</th>\n",
              "      <th>Bilirubin_total</th>\n",
              "      <th>TroponinI</th>\n",
              "      <th>Hct</th>\n",
              "      <th>Hgb</th>\n",
              "      <th>PTT</th>\n",
              "      <th>WBC</th>\n",
              "      <th>Fibrinogen</th>\n",
              "      <th>Platelets</th>\n",
              "      <th>Age</th>\n",
              "      <th>Gender</th>\n",
              "      <th>Unit1</th>\n",
              "      <th>Unit2</th>\n",
              "      <th>HospAdmTime</th>\n",
              "      <th>ICULOS</th>\n",
              "      <th>SepsisLabel</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>61.0</td>\n",
              "      <td>99.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>124.0</td>\n",
              "      <td>65.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>17.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>64.0</td>\n",
              "      <td>98.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>125.0</td>\n",
              "      <td>64.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>27.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>56.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>123.0</td>\n",
              "      <td>65.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>66.0</td>\n",
              "      <td>99.0</td>\n",
              "      <td>36.22</td>\n",
              "      <td>120.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>94.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>36.22</td>\n",
              "      <td>194.0</td>\n",
              "      <td>116.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     HR  O2Sat   Temp    SBP  ...  Unit2  HospAdmTime  ICULOS  SepsisLabel\n",
              "0  61.0   99.0  36.44  124.0  ...    1.0        -98.6       2            0\n",
              "1  64.0   98.0  36.44  125.0  ...    1.0        -98.6       3            0\n",
              "2  56.0  100.0  36.44  123.0  ...    1.0        -98.6       4            0\n",
              "3  66.0   99.0  36.22  120.0  ...    1.0        -98.6       5            0\n",
              "4  94.0  100.0  36.22  194.0  ...    1.0        -98.6       6            0\n",
              "\n",
              "[5 rows x 41 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.utils import resample\n",
        "df_majority = dataset[dataset.SepsisLabel==0]\n",
        "df_minority = dataset[dataset.SepsisLabel==1]"
      ],
      "metadata": {
        "id": "IYxEtVGOgxzT"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_minority_upsampled = resample(df_minority, \n",
        "                                 replace=True,     # sample with replacement\n",
        "                                 n_samples=57070,    # to match majority class\n",
        "                                 random_state=123) # reproducible results"
      ],
      "metadata": {
        "id": "JmZw79ieg5Vj"
      },
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_upsampled = pd.concat([df_majority, df_minority_upsampled])"
      ],
      "metadata": {
        "id": "gtAVYsXog9TT"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_upsampled.SepsisLabel.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lusO9jAehAkk",
        "outputId": "6750085c-9414-4b7e-9f32-de96aa7f480d"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    57070\n",
              "0    57070\n",
              "Name: SepsisLabel, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.pie(df_upsampled['SepsisLabel'].value_counts(), labels=['1','0'], autopct='%1.1f%%', shadow=True)\n",
        "plt.show()\n",
        "sns.countplot(x=df_upsampled['SepsisLabel'], label=\"Count\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 510
        },
        "id": "kYOTxx5thEMb",
        "outputId": "3cd82176-62af-49d5-c872-34ee76533e55"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X = df_upsampled[df_upsampled.columns[0:40]].values"
      ],
      "metadata": {
        "id": "FxkHLcSQhJxj"
      },
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y = df_upsampled[df_upsampled.columns[40:]].values\n"
      ],
      "metadata": {
        "id": "TFLQOsTwhzis"
      },
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"sepsis dimensions : {}\".format(df_upsampled.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "awELkIJBh2Jz",
        "outputId": "6118babc-9c3d-418a-a8f6-5da809fb673a"
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sepsis dimensions : (114140, 41)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"sepsis dimensions without label : {}\".format(X.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eLHYYG8Ah4Li",
        "outputId": "5ad8b448-40cf-428a-90e8-be32a42a27d7"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sepsis dimensions without label : (114140, 40)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"sepsis dimensions only label : {}\".format(Y.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TmDNHdMfh6Ri",
        "outputId": "e1d9ec49-4033-4fa6-b2f7-6da2f702bae3"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sepsis dimensions only label : (114140, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "labelencoder_Y = preprocessing.LabelEncoder()\n",
        "Y = labelencoder_Y.fit_transform(Y)"
      ],
      "metadata": {
        "id": "TOEVehzJh8_i",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6930f490-8d72-4e89-b54a-1a7ba0a50908"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/preprocessing/_label.py:115: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Y"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rBoY-PjkUnZ1",
        "outputId": "d7f4d5f7-1850-48ba-c296-adbb4e3957df"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 0, ..., 1, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.20, random_state=0)\n",
        "print(\"Training data dimensions :{}\".format(X_train.shape))\n",
        "print(\"Testing data dimensions :{}\".format(X_test.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uNNHUIm0iRU6",
        "outputId": "57d2c6c4-92b3-4253-8514-0238716b3108"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training data dimensions :(91312, 40)\n",
            "Testing data dimensions :(22828, 40)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score, log_loss\n",
        "from sklearn.neural_network import MLPClassifier"
      ],
      "metadata": {
        "id": "lN_hoUoQilEJ"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf=MLPClassifier(\n",
        "    activation='tanh',\n",
        "    solver='lbfgs',\n",
        "    early_stopping=False,\n",
        "    hidden_layer_sizes=(40,10,10,10,10, 2),\n",
        "    random_state=1,\n",
        "    batch_size='auto',\n",
        "    max_iter=13000,\n",
        "    learning_rate_init=1e-5,\n",
        "    tol=1e-4,)"
      ],
      "metadata": {
        "id": "EymmOnCqix-b"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf.fit(X_train, Y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nuVD8b4HjFyZ",
        "outputId": "150577cc-cc6e-4279-fc0e-c6d14cea4720"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:549: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "  self.n_iter_ = _check_optimize_result(\"lbfgs\", opt_res, self.max_iter)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "MLPClassifier(activation='tanh', hidden_layer_sizes=(40, 10, 10, 10, 10, 2),\n",
              "              learning_rate_init=1e-05, max_iter=13000, random_state=1,\n",
              "              solver='lbfgs')"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pickle\n"
      ],
      "metadata": {
        "id": "SRQtPw3gjpPp"
      },
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# save the model to disk\n",
        "filename = 'finalized_model2.sav'\n",
        "pickle.dump(clf, open(filename, 'wb'))"
      ],
      "metadata": {
        "id": "rswv0Lvq008y"
      },
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# load the model from disk\n",
        "loaded_model = pickle.load(open(filename, 'rb'))\n",
        "result = loaded_model.score(X_test, Y_test)\n",
        "print(result)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "92Polc9w09Lp",
        "outputId": "e60b8f51-bbc0-45db-85a8-92c20c412b0b"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9731470124408621\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix ,ConfusionMatrixDisplay\n"
      ],
      "metadata": {
        "id": "DmCinrYe1u7J"
      },
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "predicted = loaded_model.predict(X_test)\n",
        "print(predicted.shape)\n",
        "print(Y_test.shape)\n",
        "print(confusion_matrix(Y_test,predicted))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_msXZ3KY_Tks",
        "outputId": "f7bb8352-d49b-435f-cbac-875e1bd3a183"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(22828,)\n",
            "(22828,)\n",
            "[[11096   338]\n",
            " [  275 11119]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cm=confusion_matrix(Y_test,predicted)\n",
        "disp=ConfusionMatrixDisplay(confusion_matrix=cm,display_labels=clf.classes_)\n",
        "disp.plot()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "S-rbncTOAMOT",
        "outputId": "d67a0fa0-d3d2-4324-ca93-9c117a1c7f23"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://raw.githubusercontent.com/nevermind78/Capstone/main/test_D.csv"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZzoJpw9ZAbQD",
        "outputId": "2b6bb2e3-9041-476b-ae9e-e59e30e28f2f"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2021-12-09 21:33:55--  https://raw.githubusercontent.com/nevermind78/Capstone/main/test_D.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.111.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.109.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 4777460 (4.6M) [text/plain]\n",
            "Saving to: ‘test_D.csv’\n",
            "\n",
            "test_D.csv          100%[===================>]   4.56M  --.-KB/s    in 0.07s   \n",
            "\n",
            "2021-12-09 21:33:55 (61.5 MB/s) - ‘test_D.csv’ saved [4777460/4777460]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_test =pd.read_csv('test_D.csv')\n",
        "df_test=df_test.drop(\"Unnamed: 0\",axis=1)\n",
        "dataset.head()"
      ],
      "metadata": {
        "id": "3fuGgkC9YJeE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 226
        },
        "outputId": "7e6abbe5-6263-4083-eb6d-a19f8a03d9a4"
      },
      "execution_count": 62,
      "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>HR</th>\n",
              "      <th>O2Sat</th>\n",
              "      <th>Temp</th>\n",
              "      <th>SBP</th>\n",
              "      <th>MAP</th>\n",
              "      <th>DBP</th>\n",
              "      <th>Resp</th>\n",
              "      <th>EtCO2</th>\n",
              "      <th>BaseExcess</th>\n",
              "      <th>HCO3</th>\n",
              "      <th>FiO2</th>\n",
              "      <th>pH</th>\n",
              "      <th>PaCO2</th>\n",
              "      <th>SaO2</th>\n",
              "      <th>AST</th>\n",
              "      <th>BUN</th>\n",
              "      <th>Alkalinephos</th>\n",
              "      <th>Calcium</th>\n",
              "      <th>Chloride</th>\n",
              "      <th>Creatinine</th>\n",
              "      <th>Bilirubin_direct</th>\n",
              "      <th>Glucose</th>\n",
              "      <th>Lactate</th>\n",
              "      <th>Magnesium</th>\n",
              "      <th>Phosphate</th>\n",
              "      <th>Potassium</th>\n",
              "      <th>Bilirubin_total</th>\n",
              "      <th>TroponinI</th>\n",
              "      <th>Hct</th>\n",
              "      <th>Hgb</th>\n",
              "      <th>PTT</th>\n",
              "      <th>WBC</th>\n",
              "      <th>Fibrinogen</th>\n",
              "      <th>Platelets</th>\n",
              "      <th>Age</th>\n",
              "      <th>Gender</th>\n",
              "      <th>Unit1</th>\n",
              "      <th>Unit2</th>\n",
              "      <th>HospAdmTime</th>\n",
              "      <th>ICULOS</th>\n",
              "      <th>SepsisLabel</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>61.0</td>\n",
              "      <td>99.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>124.0</td>\n",
              "      <td>65.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>17.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>64.0</td>\n",
              "      <td>98.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>125.0</td>\n",
              "      <td>64.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>27.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>56.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>36.44</td>\n",
              "      <td>123.0</td>\n",
              "      <td>65.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>66.0</td>\n",
              "      <td>99.0</td>\n",
              "      <td>36.22</td>\n",
              "      <td>120.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>94.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>36.22</td>\n",
              "      <td>194.0</td>\n",
              "      <td>116.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.9</td>\n",
              "      <td>113.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>4.4</td>\n",
              "      <td>5.1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>158.0</td>\n",
              "      <td>75.91</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-98.6</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     HR  O2Sat   Temp    SBP  ...  Unit2  HospAdmTime  ICULOS  SepsisLabel\n",
              "0  61.0   99.0  36.44  124.0  ...    1.0        -98.6       2            0\n",
              "1  64.0   98.0  36.44  125.0  ...    1.0        -98.6       3            0\n",
              "2  56.0  100.0  36.44  123.0  ...    1.0        -98.6       4            0\n",
              "3  66.0   99.0  36.22  120.0  ...    1.0        -98.6       5            0\n",
              "4  94.0  100.0  36.22  194.0  ...    1.0        -98.6       6            0\n",
              "\n",
              "[5 rows x 41 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X2_test = df_test[df_test.columns[0:40]].values\n"
      ],
      "metadata": {
        "id": "K3qzaJ1zZKeT"
      },
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "Y2_test = df_test[df_test.columns[40:]].values\n",
        "Y2_test\n",
        "labelencoder_Y = preprocessing.LabelEncoder()\n",
        "Y2_test = labelencoder_Y.fit_transform(Y2_test)\n",
        "Y2_test"
      ],
      "metadata": {
        "id": "wR7Go63mZ8rZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "107d6cb2-b1f5-485b-9f96-af421b98d135"
      },
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/preprocessing/_label.py:115: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 0, ..., 0, 0, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"sepsis dimensions without label : {}\".format(X2_test.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w2UILkljaDmx",
        "outputId": "60f640e0-cb76-4bc1-9574-7f118970d688"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sepsis dimensions without label : (25210, 40)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"sepsis dimensions without label : {}\".format(Y2_test.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q5eLu1Z6aFJZ",
        "outputId": "b4ff2da5-85a7-4f6b-dd95-b77ab39ad92b"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sepsis dimensions without label : (25210,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model = pickle.load(open('finalized_model2.sav', 'rb'))\n",
        "predicted = loaded_model.predict(X2_test)\n",
        "cm=confusion_matrix(Y2_test,predicted)\n",
        "disp=ConfusionMatrixDisplay(confusion_matrix=cm,display_labels=loaded_model.classes_)\n",
        "disp.plot()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "RyuuPd03adcx",
        "outputId": "2bcaa99d-3d59-4f27-945f-a463e5b880ea"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = loaded_model.score(X2_test, Y2_test)\n",
        "print(result)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BLu7usfSb2pP",
        "outputId": "1ad3ed03-7ec5-4e7a-edd8-492618659cc5"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9255850852836176\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import precision_score, classification_report,average_precision_score, accuracy_score, balanced_accuracy_score, precision_recall_curve, f1_score, auc, roc_curve"
      ],
      "metadata": {
        "id": "smg-c8_iJ8gY"
      },
      "execution_count": 70,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "def predictors_labels_allocator(df):\n",
        "    \n",
        "    \"\"\"\n",
        "    function that allocate predictors and labels\n",
        "    \n",
        "    @param: df: Dataframe, complete dataframe\n",
        "    @return: X: Numpy Array, predictors\n",
        "             y: Numpy Array, labels\n",
        "    \"\"\"\n",
        "    col_names = df.columns\n",
        "    X = np.array(df[col_names[:-1]].values)\n",
        "    y = df[col_names[-1]].values\n",
        "    \n",
        "    return X, y"
      ],
      "metadata": {
        "id": "pju7S0hfL0dF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Xtr, ytr = predictors_labels_allocator(dataset)  \n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ub2P93cEJ8jw",
        "outputId": "116344b0-cdd1-457a-eb48-049cb65f452e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(58376, 40)"
            ]
          },
          "metadata": {},
          "execution_count": 190
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "yhat_probas = loaded_model.predict_proba(X_test)\n"
      ],
      "metadata": {
        "id": "EF8XUt1yJ8oJ"
      },
      "execution_count": 78,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pip install plot-metric"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bYTaFclqOFWq",
        "outputId": "28c1b7e2-bed3-43c1-b1d6-49d5fae8ac4c"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting plot-metric\n",
            "  Downloading plot_metric-0.0.6-py3-none-any.whl (13 kB)\n",
            "Requirement already satisfied: colorlover>=0.3.0 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (0.3.0)\n",
            "Requirement already satisfied: matplotlib>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (3.2.2)\n",
            "Requirement already satisfied: scikit-learn>=0.21.2 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (1.0.1)\n",
            "Requirement already satisfied: numpy>=1.15.4 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (1.19.5)\n",
            "Requirement already satisfied: pandas>=0.23.4 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (1.1.5)\n",
            "Requirement already satisfied: seaborn>=0.9.0 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (0.11.2)\n",
            "Requirement already satisfied: scipy>=1.1.0 in /usr/local/lib/python3.7/dist-packages (from plot-metric) (1.4.1)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=3.0.2->plot-metric) (0.11.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=3.0.2->plot-metric) (3.0.6)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=3.0.2->plot-metric) (1.3.2)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=3.0.2->plot-metric) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.23.4->plot-metric) (2018.9)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.1->matplotlib>=3.0.2->plot-metric) (1.15.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.21.2->plot-metric) (3.0.0)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.21.2->plot-metric) (1.1.0)\n",
            "Installing collected packages: plot-metric\n",
            "Successfully installed plot-metric-0.0.6\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%matplotlib inline\n",
        "%config InlineBackend.figure_format = 'retina'\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib\n",
        "import matplotlib.pyplot as plt\n",
        "import os\n",
        "import random\n",
        "from plot_metric.functions import BinaryClassification"
      ],
      "metadata": {
        "id": "OSHE96vBNGoe"
      },
      "execution_count": 73,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_ROC(y_test, y_score, n_classes=2):\n",
        "    # Compute ROC curve and ROC area for each class\n",
        "    fpr = dict()\n",
        "    tpr = dict()\n",
        "    roc_auc = dict()\n",
        "    fpr['positive'], tpr['positive'], _ = roc_curve(y_test, y_score)\n",
        "    roc_auc['positive'] = auc(fpr['positive'], tpr['positive'])\n",
        "\n",
        "    # Compute micro-average ROC curve and ROC area\n",
        "    fpr[\"micro\"], tpr[\"micro\"], _ = roc_curve(y_test.ravel(), y_score.ravel())\n",
        "    roc_auc[\"micro\"] = auc(fpr[\"micro\"], tpr[\"micro\"])\n",
        "    \n",
        "    plt.figure()\n",
        "    lw = 2\n",
        "    plt.plot(fpr['positive'], tpr['positive'], color='darkorange',\n",
        "             lw=lw, label='ROC curve (area = %0.2f)' % roc_auc['positive'])\n",
        "    plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n",
        "    plt.xlim([0.0, 1.0])\n",
        "    plt.ylim([0.0, 1.05])\n",
        "    plt.xlabel('False Positive Rate')\n",
        "    plt.ylabel('True Positive Rate')\n",
        "    plt.title('Receiver operating characteristic curve')\n",
        "    plt.legend(loc=\"lower right\")\n",
        "    plt.show()"
      ],
      "metadata": {
        "id": "DLS7Y55GNtvp"
      },
      "execution_count": 74,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y_test"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZlwNTDMbOPQX",
        "outputId": "ee8c49e9-44ba-4eb7-95fa-dc9e90abf5f8"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 0, 0, ..., 1, 1, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u4jepGk8OsAu",
        "outputId": "d9511056-00e3-41af-d43a-755a0d26873f"
      },
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 9.9000e+01,  9.4000e+01,  3.7560e+01, ...,  0.0000e+00,\n",
              "        -2.0000e-02,  2.0000e+01],\n",
              "       [ 6.8000e+01,  1.0000e+02,  3.5940e+01, ...,  0.0000e+00,\n",
              "        -6.5700e+00,  1.4000e+01],\n",
              "       [ 6.2000e+01,  9.6000e+01,  3.5390e+01, ...,  0.0000e+00,\n",
              "        -2.6118e+02,  2.8000e+01],\n",
              "       ...,\n",
              "       [ 8.6000e+01,  9.7000e+01,  3.6890e+01, ...,  0.0000e+00,\n",
              "        -1.4006e+02,  3.0000e+00],\n",
              "       [ 9.6000e+01,  1.0000e+02,  3.6670e+01, ...,  0.0000e+00,\n",
              "        -3.0000e-02,  1.9200e+02],\n",
              "       [ 1.0700e+02,  9.6000e+01,  3.8110e+01, ...,  0.0000e+00,\n",
              "        -1.0000e-02,  5.5000e+01]])"
            ]
          },
          "metadata": {},
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_ROC(Y_test,yhat_probas[:,1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 294
        },
        "id": "M-FR_A86OuDm",
        "outputId": "4f96cb18-03b2-437e-93aa-a1b31379f575"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "image/png": {
              "width": 393,
              "height": 277
            },
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "O9PY3EtDYCBT"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}