Phishing_Random_Forest.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2lJWHerO1bKp"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "#from sklearn.externals import joblib"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = pd.read_csv(\"training.csv\")"
      ],
      "metadata": {
        "id": "BJzYdNLl1cez"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x = dataset.iloc[ : , :-1].values\n",
        "y = dataset.iloc[:, -1:].values"
      ],
      "metadata": {
        "id": "FijgAsAf1rBT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_train, x_test, y_train, y_test = train_test_split(x,y,test_size = 0.25, random_state =0 )"
      ],
      "metadata": {
        "id": "cCS-ICul1zET"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "parameters = [{'n_estimators': [100, 800],\n",
        "    'max_features': ['sqrt', 'log2'],\n",
        "    'criterion' :['gini', 'entropy']}]"
      ],
      "metadata": {
        "id": "M-v3j95z12h7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search = GridSearchCV(RandomForestClassifier(),  parameters,cv =5, n_jobs= -1)\n",
        "grid_search.fit(x_train, y_train)\n",
        "#printing best parameters\n",
        "print(\"Best Accurancy =\" +str( grid_search.best_score_))\n",
        "print(\"best parameters =\" + str(grid_search.best_params_))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9Dch0OF92ItV",
        "outputId": "fb3dad7d-1b99-4e69-fe38-e805957fb306"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/model_selection/_search.py:926: 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",
            "  self.best_estimator_.fit(X, y, **fit_params)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best Accurancy =0.9642643291493813\n",
            "best parameters ={'criterion': 'entropy', 'max_features': 'sqrt', 'n_estimators': 800}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "classifier = RandomForestClassifier(n_estimators = 200, criterion = \"gini\", max_features = 'log2',  random_state = 0)\n",
        "classifier.fit(x_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XHXzijLl2QmB",
        "outputId": "e298bd6c-32f2-4c26-d9a9-197a67c073f3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:2: 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",
            "  \n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(max_features='log2', n_estimators=200, random_state=0)"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = classifier.predict(x_test)"
      ],
      "metadata": {
        "id": "vn8-QGL72niN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data=pd.read_csv('testing.csv')"
      ],
      "metadata": {
        "id": "IWtxtDZa3tmN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "preds=classifier.predict(test_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l5_N86k732ll",
        "outputId": "b00a4f7b-cf86-4c9f-edce-a47da0615709"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/base.py:444: UserWarning: X has feature names, but RandomForestClassifier was fitted without feature names\n",
            "  f\"X has feature names, but {self.__class__.__name__} was fitted without\"\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data['Result']=preds\n",
        "submission=test_data.loc[:,['key','Result']]\n",
        "submission.to_csv('accuracycheck1.csv',index=False)"
      ],
      "metadata": {
        "id": "6WaqrfY-4D61"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "HZngSw7d4Hyl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "byOx5oqg35Kk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "cm = confusion_matrix(y_test, y_pred)\n",
        "print(cm)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gypuFpRS23s4",
        "outputId": "a5d6c376-e1c7-48a6-e253-0bf404bca1ad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[ 950   48]\n",
            " [  30 1211]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "names = dataset.iloc[:,:-1].columns\n",
        "importances =classifier.feature_importances_\n",
        "sorted_importances = sorted(importances, reverse=True)\n",
        "indices = np.argsort(-importances)\n",
        "var_imp = pd.DataFrame(sorted_importances, names[indices], columns=['importance'])"
      ],
      "metadata": {
        "id": "1KqXYYkv296U"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.title(\"Variable Importances\")\n",
        "plt.barh(np.arange(len(names)), sorted_importances, height = 0.7)\n",
        "plt.yticks(np.arange(len(names)), names[indices], fontsize=7)\n",
        "plt.xlabel('Relative Importance')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "vdmjcRIm3EqZ",
        "outputId": "bb7b836f-1370-4cb0-a927-7c7e086c06d5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "f8nx9nfK3gqe"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}