Build_Machine_Learning_Pipelines.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Build Machine Learning Pipelines.ipynb",
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyO09LGomLH89dzauw10sQmb",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/hussain0048/Machine-Learning/blob/master/Build_Machine_Learning_Pipelines.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Import Necessary Libraries**"
      ],
      "metadata": {
        "id": "ceFopDVNcbe_"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "D0n2uYOTcWW9"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.preprocessing import RobustScaler\n",
        "from sklearn.decomposition import PCA"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Dataset**"
      ],
      "metadata": {
        "id": "OkHSgdnKeGp2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# read in the iris data\n",
        "from sklearn.datasets import load_iris\n",
        "iris = load_iris()\n",
        "\n",
        "# create X (features) and y (response)\n",
        "X = iris.data\n",
        "y = iris.target"
      ],
      "metadata": {
        "id": "bX7BZLHGeOGt"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Split Dataset into Training and Testing Set**"
      ],
      "metadata": {
        "id": "6-BAJe0NeXgC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn import preprocessing"
      ],
      "metadata": {
        "id": "vpl-tg3CejB1"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X,y)"
      ],
      "metadata": {
        "id": "L5Qx_7CCenus"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Machine Learning Model Pipeline"
      ],
      "metadata": {
        "id": "2MySkg4lfvas"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "# instantiate the model\n",
        "logreg = LogisticRegression(solver='liblinear')\n",
        "from sklearn.svm import SVC\n",
        "#from sklearn.neighbors import kneighborsClassifiers\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "\n",
        "model_pipline =[]\n",
        "model_pipline.append(LogisticRegression (solver='liblinear'))\n",
        "model_pipline.append(SVC())\n",
        "model_pipline.append(KNeighborsClassifier())\n",
        "model_pipline.append(DecisionTreeClassifier())\n",
        "model_pipline.append(RandomForestClassifier())\n",
        "model_pipline.append(GaussianNB())"
      ],
      "metadata": {
        "id": "Q1vCuLnpgbRk"
      },
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Model Evaluation**"
      ],
      "metadata": {
        "id": "WRxTh7A9sBAB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn import metrics\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import roc_curve\n"
      ],
      "metadata": {
        "id": "esj7yOJOsSEA"
      },
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_list=['logistic Regression','SVM','KNN', 'DT', 'RF','NB']\n",
        "acc_list = []\n",
        "auc_list = []\n",
        "cm_list = []"
      ],
      "metadata": {
        "id": "i62eDHaCs67P"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for model in model_pipline:\n",
        "  model.fit(X_train,y_train)\n",
        "  y_pred = model.predict(X_test)\n",
        "  acc_list.append(metrics.accuracy_score(y_test, y_pred))\n",
        "  #fpr, tpr, _thresholds = metrics.roc_curve(y_test, y_pred)\n",
        "  fpr, tpr, _thresholds = roc_curve(y_test, y_pred, pos_label=1)\n",
        "  auc_list.append (round(metrics.auc(fpr, tpr),2))\n",
        "  cm_list.append(confusion_matrix(y_test, y_pred))"
      ],
      "metadata": {
        "id": "iefHX_4Lt7y9"
      },
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Confusion matrix**"
      ],
      "metadata": {
        "id": "YQyzrTtmM3YN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns \n",
        "fig = plt.figure (figsize = (18,10))\n",
        "for i in range(len(cm_list)):\n",
        "  cm = cm_list[i]\n",
        "  model = model_list[i]\n",
        "  sub = fig.add_subplot(2, 3, i+1).set_title(model)\n",
        "  cm_plot = sns.heatmap(cm, annot= True, cmap ='Blues_r')\n",
        "  cm_plot.set_xlabel('predicted Values')\n",
        "  #cm_plot.set_ylablel('Actual Values')\n",
        "  cm_plot.set_ylabel('Actual Values')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 641
        },
        "id": "YTxbYDIZKixG",
        "outputId": "1bccf351-1943-4377-8c9c-2dfb2cb63cbd"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x720 with 12 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Accuracy**"
      ],
      "metadata": {
        "id": "R6Jh5jAuM8Bo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#result_df= pd.DataFrame({'Model':model_list, 'Accuracy': acc_list, 'AUC': auc_list})\n",
        "result_df =pd.DataFrame({'Model':model_list, 'Accuracy': acc_list,'AUC': auc_list })\n",
        "result_df"
      ],
      "metadata": {
        "id": "zgdCSTfwx2IE"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Save Model**"
      ],
      "metadata": {
        "id": "zRuyVCVgqziA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import joblib\n",
        "joblib.dump(model, 'pipe.joblib')\n"
      ],
      "metadata": {
        "id": "6SjX9Vv0rAib"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "For more detail please visite [coursesteach.com](https://coursesteach.com/)"
      ],
      "metadata": {
        "id": "RKEAa9qqwmMd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **References**"
      ],
      "metadata": {
        "id": "0d0iphaCNHyx"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "[1-Top 6 Machine Learning Algorithms for Classification](https://towardsdatascience.com/top-machine-learning-algorithms-for-classification-2197870ff501)"
      ],
      "metadata": {
        "id": "MFvhgmpmNL-q"
      }
    }
  ]
}