Logistic_Regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Logistic Regression.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyMmchgj4I9iAjIikMDVfvqJ",
      "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/Jacqueline45/EE559_Project/blob/main/Logistic_Regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip /content/EE559_Project-main.zip"
      ],
      "metadata": {
        "id": "PxjyMdkEKG1c"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ],
      "metadata": {
        "id": "SLnLo2mgKHHp"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fires_train = pd.read_csv(\"/content/EE559_Project-main/datasets/algerian_fires_train.csv\")\n",
        "fires_test = pd.read_csv(\"/content/EE559_Project-main/datasets/algerian_fires_test.csv\")\n",
        "\n",
        "fires_trainx = fires_train.iloc[:,1:-1]\n",
        "fires_trainy = fires_train.iloc[:,-1]\n",
        "fires_testx = fires_test.iloc[:,1:-1]\n",
        "fires_testy = fires_test.iloc[:,-1]\n",
        "\n",
        "scalar = MinMaxScaler()\n",
        "\n",
        "fires = scalar.fit_transform(fires_trainx)\n",
        "firestst = scalar.fit_transform(fires_testx)"
      ],
      "metadata": {
        "id": "hFxRpmZN9ezY"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def metrics(true_labels, pred_labels, plot_title, work='test'):\n",
        "    cf_matrix = confusion_matrix(true_labels, pred_labels)\n",
        "    TP, TN, FP, FN = cf_matrix[1][1], cf_matrix[0][0], cf_matrix[0][1], cf_matrix[1][0]\n",
        "    Recall = TP/(TP+FN)\n",
        "    Precision = TP/(TP+FP)\n",
        "    F1_score = 2*Recall*Precision/(Recall+Precision)\n",
        "    Accuracy = (TP+TN)/true_labels.shape[0]\n",
        "    \n",
        "    if work == 'test':\n",
        "        ax = sns.heatmap(cf_matrix, annot=True, cmap='Blues')\n",
        "\n",
        "        ax.set_title(plot_title)\n",
        "        ax.set_xlabel('Predicted Values')\n",
        "        ax.set_ylabel('Actual Values ')\n",
        "\n",
        "        ax.xaxis.set_ticklabels(['0','1'])\n",
        "        ax.yaxis.set_ticklabels(['0','1'])\n",
        "\n",
        "        plt.savefig('/content/EE559_Project-main/cf_matrix_plots/'+plot_title+'.png')\n",
        "\n",
        "        return F1_score, Accuracy\n",
        "\n",
        "    else:\n",
        "        return F1_score, Accuracy, TP, TN, FP, FN"
      ],
      "metadata": {
        "id": "meS4gtUbM8ZK"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import statsmodels.api as sm\n",
        "import pandas as pd\n",
        "\n",
        "X = sm.add_constant(fires_trainx)\n",
        "log_reg = sm.Logit(fires_trainy, X)\n",
        "log1 = log_reg.fit(method = \"cg\")\n",
        "print(log1.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l6zEdKb_Hkaf",
        "outputId": "99215ee4-e54b-4229-ded6-c631eb213e75"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Warning: Maximum number of iterations has been exceeded.\n",
            "         Current function value: 0.245380\n",
            "         Iterations: 35\n",
            "         Function evaluations: 66\n",
            "         Gradient evaluations: 66\n",
            "                           Logit Regression Results                           \n",
            "==============================================================================\n",
            "Dep. Variable:                Classes   No. Observations:                  184\n",
            "Model:                          Logit   Df Residuals:                      174\n",
            "Method:                           MLE   Df Model:                            9\n",
            "Date:                Sun, 01 May 2022   Pseudo R-squ.:                  0.6291\n",
            "Time:                        20:30:24   Log-Likelihood:                -45.150\n",
            "converged:                      False   LL-Null:                       -121.73\n",
            "Covariance Type:            nonrobust   LLR p-value:                 1.956e-28\n",
            "===============================================================================\n",
            "                  coef    std err          z      P>|z|      [0.025      0.975]\n",
            "-------------------------------------------------------------------------------\n",
            "const          -0.0270      1.395     -0.019      0.985      -2.761       2.707\n",
            "Temperature    -0.0085      0.025     -0.346      0.729      -0.057       0.040\n",
            "RH             -0.0156      0.012     -1.360      0.174      -0.038       0.007\n",
            "Ws             -0.0281      0.044     -0.632      0.527      -0.115       0.059\n",
            "Rain           -0.1213      0.176     -0.691      0.489      -0.465       0.223\n",
            "FFMC           -0.0023      0.011     -0.203      0.839      -0.024       0.019\n",
            "DMC            -0.0076      0.043     -0.178      0.859      -0.091       0.076\n",
            "DC              0.0112      0.010      1.079      0.280      -0.009       0.031\n",
            "ISI             0.5235      0.120      4.364      0.000       0.288       0.759\n",
            "BUI             0.0167      0.045      0.374      0.708      -0.071       0.104\n",
            "===============================================================================\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:512: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  \"Check mle_retvals\", ConvergenceWarning)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Taking all features\n",
        "\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "import statsmodels.api as sm\n",
        "from sklearn.metrics import confusion_matrix,roc_curve, auc,roc_auc_score\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "logreg = LogisticRegression()\n",
        "logreg.fit(fires_trainx, fires_trainy)\n",
        "Y_pred = logreg.predict(fires_trainx)\n",
        "val=logreg.score(fires_trainx,fires_trainy)\n",
        "print(\"Accuracy for train:\",val)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_trainy, Y_pred)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label');\n",
        "plt.xlabel('Predicted label');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "TvvsnstcRvPe",
        "outputId": "fed51fc3-fe62-43fa-f8f9-d67dc688e55e"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for train: 0.9239130434782609\n",
            "--------------------------\n",
            "[[ 63   6]\n",
            " [  8 107]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "F1_score, Accuracy = metrics(fires_trainy,Y_pred, \"Logistic_reg_train\")\n",
        "print(\"Train Logistic regression F1_score=\", F1_score, \"Train Logistic regression Accuracy=\", Accuracy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "X26AvnFROeGy",
        "outputId": "2a0361a4-7a73-40ed-a115-68fe8b4d6ea7"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Logistic regression F1_score= 0.9385964912280702 Train Logistic regression Accuracy= 0.9239130434782609\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Taking all features\n",
        "Y_predtst = logreg.predict(fires_testx)\n",
        "val=logreg.score(fires_testx,fires_testy)\n",
        "print(\"Accuracy for test:\",val)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_testy, Y_predtst)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual Test label');\n",
        "plt.xlabel('Predicted Test label');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "Y5AgZmpxSkVP",
        "outputId": "dbd4668d-0b60-4121-a1c8-efd48f8abe6e"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for test: 0.8666666666666667\n",
            "--------------------------\n",
            "[[33  4]\n",
            " [ 4 19]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAATYAAAEGCAYAAADvxrkEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAYMUlEQVR4nO3de5wcZZ3v8c93ZnLDhMsYyAkXVxIQNrJLkEhY5bjcQXFfgsZLdBX3sAY9i1w2eEDXg7DHl4KiLpoDMhgkKiCwhAOyKsSQmCAaEiAgCSxBjUrIBdYgJCKY5Hf+6Bpo4kx3daUvVTXfN696TXX19FO/JuTLU/VUPaWIwMysTLo6XYCZWbM52MysdBxsZlY6DjYzKx0Hm5mVTk+nCxjMqEPO8HBtgWxcOqvTJVgGI3vQjny+kb+nzz8wa4f21Qj32MysdHLbYzOzAlA++0YONjPLrqu70xUMyMFmZtmpbafNGuJgM7PsfChqZqXjHpuZlY57bGZWOu6xmVnpeFTUzErHh6JmVjo+FDWz0nGPzcxKx8FmZqXT7cEDMysbn2Mzs9LxoaiZlY57bGZWOu6xmVnp5LTHls+4NbNi6OpOv9QhaaSkeyU9KGmFpIuS7ftKWiLpcUk3SBpet6wmfDUzG6rUlX6p7wXg6Ig4GJgMnCjpcOAS4CsRsR+wETitXkMONjPLTkq/1BEVm5KXw5IlgKOBf0+2zwFOrteWg83MsmugxyZphqRlVcuMP2tO6pa0HNgAzAN+ATwTEVuSX3kC2KteWR48MLPsGhgVjYg+oK/O72wFJkvaFbgFODBLWQ42M8uuRfOxRcQzkhYAfwPsKqkn6bXtDaypW1ZLqjKzoaGJ59gk7Z701JA0CjgOeARYAExLfu1U4NZ6bbnHZmbZNfcC3fHAHEndVDpdN0bE7ZJWAt+V9FngAWB2vYYcbGaWXRMv0I2Ih4BDBtj+S+CwRtpysJlZZsrpnQcONjPLzMFmZqWjLgebmZWMe2xmVjoONjMrHQebmZVPPnPNwWZm2bnHZmal09WVz7syHWxmlpl7bGZWPvnMNQebmWXnHpuZlY6DzcxKx7dUmVnpuMdmZqXjYDOz0nGwmVnpONjMrHzymWsONjPLzrdUmVnp+FDUzMonn7nmYGulEcN7+NHssxk+vIee7m5u+dEDfPbr3+eKz7yfN0x6DUI8/psNfOSCb7P5+Rc7Xa4NYuvWrUx/z7vYY9w4Zl1+ZafLyRX32IagF17cwokzvsrm51+kp6eLu67+Z+78yUr+16VzeW7zHwG4ZOY7+dj7/pZLvzmvw9XaYK799reYMGEimzZv6nQpuZPXYMvnmb8S6e+JDevppqenm4h4KdQARo4YRkR0qjyrY/26dSxetJBT3jWt06XkkqTUSzu1rMcm6UDgHcBeyaY1wG0R8Uir9plHXV3inuvOY+I+u3PlDYtY+vCvAbjywr/nhCMm8egv13H+l+d2uEobzBcu/hznzPwEmzdv7nQpuZTXe0Vb0mOTdB7wXSqnFu9NFgHXSzq/xudmSFomadmWp1e0orS227YtOPx9F7PfCZ9mykF/waSJ4wE4/cLvMOH4f+HRX61j2vGHdrhKG8iPFy6gt7eXSa8/qNOl5FZee2ytOhQ9DXhjRFwcEd9JlouBw5L3BhQRfRExJSKm9Ix9fYtK64zfb3qeHy97jOPfNOmlbdu2BTfdcR8nHzO5g5XZYJY/cD8LF97FW487mvPO/WeWLvkZnzzv3E6XlStDLdi2AXsOsH188t6QMHa30ewyehRQOZd2zNQDeezX65mwz9iXfuftf/vXPLZ6fadKtBrOOmcm8+5axA/m3cUll36ZN049nM9fcmmny8oVKf1Svy3tI2mBpJWSVkg6K9l+oaQ1kpYny9vqtdWqc2xnA/MlrQJ+m2x7DbAfcEaL9pk7/23szlz1rx+ku6uLri5x87z7+cHiFcy/+mzGvGoUEvz8sTWc+bkbOl2qWSZN7oltAWZGxP2SxgD3Seq/XOArEZH6/ypq1YicpC4qh57VgwdLI2Jrms+POuQMDxUWyMalszpdgmUwsmfHLrE94Lw7Uv89/c9LTmhoX5JuBWYBbwY2NRJsLRsVjYhtwM9a1b6ZdV4jHTZJM4AZVZv6IqJvkN99LXAIsIRKsJ0h6UPAMiq9uo219uXr2Mwss64upV6qBweTZbBQGw3cDJwdEc8CVwATgcnAWuBL9erynQdmllmzBzslDaMSatdGxFyAiFhf9f5VwO312nGwmVlmzRw8UKWx2cAjEfHlqu3jI2Jt8vIU4OF6bTnYzCyzJvfY3gx8EPi5pOXJtk8B0yVNBgJYDZxeryEHm5ll1syJJiPibgaeCOn7jbblYDOzzHI6uYeDzcyyy+u0RQ42M8ssp7nmYDOz7NxjM7PSyWmuDR5skr5GZXh1QBFxZksqMrPC6MrpRJO1emzL2laFmRVS4Q5FI2JO9WtJO0XEH1pfkpkVRU5zrf5N8JL+RtJK4NHk9cGSLm95ZWaWe0WeQfffgBOA/wKIiAeBt7SyKDMrhmbOoNtMqUZFI+K32yVuqskizazcijh40O+3kt4ERDKlyFnAkHqEnpkNrHCDB1U+ClxGZYrvJ4E7gH9qZVFmVgyFDbaIeBr4QBtqMbOCyWmupRoVnSDpe5KekrRB0q2SJrSjODPLtyKPil4H3EjlmaB7AjcB17eyKDMrhryOiqYJtp0i4tsRsSVZvgOMbHVhZpZ/jTzMpZ1q3Svam6z+QNL5wHep3Dv6XjLMaGlm5dOV05NstQYP7qMSZP2VV88zHsAnW1WUmRVDTnOt5r2i+7azEDMrnsJe7gEg6SBgElXn1iLiW60qysyKIac3HtQPNkmfAY6kEmzfB94K3A042MyGuLzeUpVmVHQacAywLiL+ATgY2KWlVZlZIaiBf9opzaHo8xGxTdIWSTsDG4B9WlyXmRVATjtsqYJtmaRdgauojJRuAn7a0qrMrBAKO3gQEf8zWf26pB8CO0fEQ60ty8yKIKe5VvMC3TfUei8i7m9NSWZWFEW8QPdLNd4L4Ogm12JmBZPXUdFaF+ge1c5CzKx4ctphS3W5h5nZgLqk1Es9kvaRtEDSSkkrJJ2VbO+VNE/SquTnbnXrasJ3M7MhSg0sKWwBZkbEJOBw4J8kTQLOB+ZHxP7A/OR1TQ42M8usmRNNRsTa/kHJiHiOyrNV9gLeAfQ/53gOcHK9ttLMoDs/zTYzG3q6lH6RNEPSsqplxmDtSnotcAiwBBgXEWuTt9YB4+rVVetyj5HATsDY5Ji2P3J3ppKiZjbENTIqGhF9QF+935M0GrgZODsinq3u7UVESIp6bdS63ON04Gwq04Hfx8vB9iwwq17DZlZ+zb7zIHnE583AtRExN9m8XtL4iFgraTyV2zprGvRQNCIuS+ZkOzciJkTEvslycEQ42MysoUPRelRJydnAIxHx5aq3bgNOTdZPBW6tW1eK2tdJGpPs+NOS5ta6K8HMho4mP6XqzcAHgaMlLU+WtwEXA8dJWgUcm7yuKc1N8P87Im6SdETS6BeBK4CpaSo1s/Jq5oFoRNxdo8ljGmkrTY9ta/LzJKAvIv4DGN7ITsysnLq7lHpppzQ9tjWSrgSOAy6RNAJf/2Zm5HfaojQB9R7gDuCEiHgG6AU+0dKqzKwQCvvA5Ij4A5Xh1SOSTVuAVa0sysyKoZn3ijZT2oe5TAEOAL4JDAO+Q2UEw8yGsJweiaY6x3YKlVsb+u/herL/8o9W2rjUl8oVyZm3rOh0CZZB37tfv0Ofz+s5tjTB9mL1bQySXtXimsysILpzGmyDnmOT9Llk9cZkVHRXSR8BfkTlwS5mNsQ1886DZqrVYzsR+FREXCrpOCr3iB4AXBAR89pSnZnlWk5nBq8ZbN1Vs3rclyxAZUbLiPhdq4szs3wr4jm2A3l5Vo/qaUL6X09oYV1mVgBF7LGtjIhD2laJmRVOTjtsqUZFzcwG1JPTZKsVbJe1rQozK6Sc5lrN54pe08Y6zKyAivgkeDOzmnKaaw42M8uucKOikr7GKy/zeIWIOLMlFZlZYbR7Asm0avXYlrWtCjMrpJzmWs3BgzmDvWdmBqCmPvWgedLMx7Y7cB4wCRjZvz0ijm5hXWZWAHntsaWZGvxa4BFgX+AiYDWwtIU1mVlB5HV2jzTB9uqImA38KSJ+HBH/A3Bvzcya/VzRpklzucefkp9rJZ0EPEnlgS5mNsR15/R5dWmC7bOSdgFmAl8DdgbOaWlVZlYIhb3zICJuT1Z/DxzV2nLMrEjyOniQZlT0mwxwoW5yrs3MhrCcdthSHYreXrU+kspTq55sTTlmViRdRb2OLSJurn4t6Xrg7pZVZGaF0cwem6SrgbcDGyLioGTbhcBHgKeSX/tURHy/XltZxjT2B/bI8DkzK5meLqVeUriGykOktveViJicLHVDDdKdY3uOV55jW0flTgQzG+Ka2WOLiEWSXtuMttIcirb8qe9mVkyNXO4haQYwo2pTX0T0pfjoGZI+RGVijpkRsbFuXSmKmZ9mm5kNPVL6JSL6ImJK1ZIm1K4AJgKTgbXAl9LUVWs+tpHATsDYqueLQuUC3b3SNG5m5dbqGw8iYn3/uqSreOVVGoOqdSh6OnA2sCcvP18UKk+En5WtTDMrk1bfeSBpfESsTV6eAjyc5nO15mO7DLhM0scj4mtNqNHMSqaZwZZcSnYklaPEJ4DPAEdKmkxlAHM1lQ5XXWku0N0madeIeCbZ+W7A9Ii4PEPtZlYizeyvRcT0ATbPztJWmkPkj/SHWrLzjVQumDOzIa6RwYN2StNj65akiAgASd3A8NaWZWZF0O551tJKE2w/BG6QdGXy+vRkm5kNcTmdji1VsJ1H5aK6jyWv5wFXtawiMyuMvM7HVjdwI2JbRHw9IqZFxDRgJZUJJ81siCvy1OBIOgSYDrwH+BUwt5VFmVkxFO5QVNLrqITZdOBp4AZAEeFZdM0MKObgwaPAYuDtEfE4gCQ/68DMXpLPWKvdk3wnlZtOF0i6StIx5Pd7mFkHdEupl3YaNNgi4v9FxPuAA4EFVO4b3UPSFZKOb1eBZpZfeb1AN82o6OaIuC4i/g7YG3gATzRpZoAa+KedGhrUiIiNyZxKx7SqIDMrjrz22FJd7mFmNpDCPqXKzGwwOb3aw8FmZtnl9ZYqB5uZZZbuqXrt52Azs8zaPdqZloPNzDLL6ZGog63dtm7dyvT3vIs9xo1j1uVX1v+AtdWpU/bkr8aP4bkXtnDRnb8AYO9dRvCBQ/dkZE8XT2/+E7OXPMEft2zrcKX5kNceW15vzi+ta7/9LSZMmNjpMmwQ96x+hq8u/vUrtn1oyl7c8tB6LrrzFyxf8yzHHzC2Q9XlT5fSL22tq727G9rWr1vH4kULOeVd0zpdig1i1dN/YPOLW1+xbdyY4Tz29B8AWLl+E2/Ye0wnSsulLin10ta62rq3Ie4LF3+Oc2Z+gq4u/2svkid//wKT96yE2aF770LvqGEdrig/1MDSTm3/GybpH2q8N0PSMknLZl/V186yWu7HCxfQ29vLpNcf1OlSrEFzlq3hyIm9/MuxExg5rIst26LTJeVGXntsnRg8uAj45kBvREQf0Afwxy2U6r+e5Q/cz8KFd3H34kW88MILbN68iU+edy6fv+TSTpdmdax77kX+LTnvtsfo4fzV+NEdrig/8jl00KJgk/TQYG8B41qxz7w765yZnHXOTACW3ruEOddc7VAriDEjunnuha0IOOkvd2fRLzZ2uqT8yGmytarHNg44Adj+vwAB97Ron2Y77B+n7s0Bu+/E6BE9XHLS67htxQZG9HRx1H69ANy/5ll+svqZOq0MHUPtlqrbgdERsXz7NyQtbNE+C+ONh03ljYdN7XQZNoBvLHliwO13Pf67NldSDPmMtRYFW0ScVuO997din2bWATlNNt95YGaZ+c4DMyudZs6gK+lqSRskPVy1rVfSPEmrkp+7panLwWZmmTX5At1rgBO323Y+MD8i9gfmJ6/rcrCZWWaSUi/1RMQiYPtRmncAc5L1OcDJaepysJlZZo0cilbfWZQsM1LsYlxErE3W15HyOlgPHphZZo0MHVTfWZRFRISkVHckucdmZtm1/i749ZLGAyQ/N6T5kIPNzDJrwwOTbwNOTdZPBW5N8yEHm5ll1uTLPa4HfgocIOkJSacBFwPHSVoFHJu8rsvn2Mwss2beKhoR0wd565hG23KwmVlmeb3zwMFmZpnldHIPB5uZZZfTXHOwmdkOyGmyOdjMLLOhNtGkmQ0B+Yw1B5uZ7YicJpuDzcwy8+UeZlY6OT3F5mAzs+xymmsONjPLLs0Ekp3gYDOzzHKaaw42M8sup7nmYDOzHZDTZHOwmVlmvtzDzErH59jMrHS6HGxmVj75TDYHm5ll5kNRMyudnOaag83MsnOPzcxKx7dUmVnp5DPWHGxmtgNy2mFzsJlZdr7zwMzKJ5+55mAzs+xymmsONjPLzo/fM7PSaXauSVoNPAdsBbZExJQs7TjYzCxvjoqIp3ekAQebmWWW0yNRujpdgJkVlxr5R5ohaVnVMmOAJgO4U9J9g7yfintsZpZZIz22iOgD+ur82hERsUbSHsA8SY9GxKJG63KPzcwyk9IvaUTEmuTnBuAW4LAsdTnYzCyzRg5F67YlvUrSmP514Hjg4Sx1+VDUzDJr8uDBOOCWZMaQHuC6iPhhloYcbGaWWTNzLSJ+CRzcjLYcbGaWXU4v93CwmVlmeb2lShHR6RqGHEkzkqFvKwD/eRWPR0U7I/OFh9YR/vMqGAebmZWOg83MSsfB1hk+X1Ms/vMqGA8emFnpuMdmZqXjYDOz0nGwtZGkEyX9p6THJZ3f6XqsNklXS9ogKdON2NY5DrY2kdQN/F/grcAkYLqkSZ2tyuq4Bjix00VY4xxs7XMY8HhE/DIiXgS+C7yjwzVZDckEh7/rdB3WOAdb++wF/Lbq9RPJNjNrMgebmZWOg6191gD7VL3eO9lmZk3mYGufpcD+kvaVNBx4H3Bbh2syKyUHW5tExBbgDOAO4BHgxohY0dmqrBZJ1wM/BQ6Q9ISk0zpdk6XjW6rMrHTcYzOz0nGwmVnpONjMrHQcbGZWOg42MysdB1uOSNoqabmkhyXdJGmnHWjrGknTkvVv1LrhXtKRkt6UYR+rJY3dbtuS5Dv8RtJTyfpySa9N2eZkSW+rUeftdT7/YUmz0n6H5DN/9j2s2Bxs+fJ8REyOiIOAF4GPVr8pKdNzYCPiHyNiZY1fORJoONgG2dfUiJgMXADckHyfyRGxOmUTk4EBg80sLQdbfi0G9kt6KYsl3QaslNQt6YuSlkp6SNLpAKqYlcz39iNgj/6GJC2UNCVZP1HS/ZIelDQ/6Ul9FDgn6Vn9d0m7S7o52cdSSW9OPvtqSXdKWiHpG6R8DrikiZJ+KOm+5LscmGx/d9I7fVDSouSOjH8F3pvU8t4abR4m6aeSHpB0j6QDqt7eJ/nOqyR9puozfy/p3qTtK5OppKyMIsJLThZgU/KzB7gV+BiV3tRmYN/kvRnAp5P1EcAyYF/gncA8oBvYE3gGmJb83kJgCrA7lRlG+tvqTX5eCJxbVcd1wBHJ+muAR5L1rwIXJOsnAQGMHeS7fBiYlazPB/ZP1qcCdyXrPwf2StZ33f5zA7R5JHB7sr4z0JOsHwvcXPX5tcCrgVHAw8l3/0vge8Cw5PcuBz6UrK8e7Ht4KeaS6dDGWmaUpOXJ+mJgNpVDxHsj4lfJ9uOBv+4/fwbsAuwPvAW4PiK2Ak9KumuA9g8HFvW3FRGDzTV2LDBJeqlDtrOk0ck+3pl89j8kbaz3hZLPvQm4qaq9EcnPnwDXSLoRmFuvre3sAsyRtD+VgB1W9d68iPivZP9zgSOALcChwNKkjlHAhgb3aQXhYMuX56NyfuolyV/CzdWbgI9HxB3b/V4zz0t1AYdHxB8HqCVLW89s/70AIuKjkqZS6f3dJ+nQBtr9P8CCiDglOZxeWN309rui8u9tTkR8soF9WEH5HFvx3AF8TNIwAEmvk/QqYBGVc1PdksYDRw3w2Z8Bb5G0b/LZ3mT7c8CYqt+7E/h4/wtJ/aG0CHh/su2twG71io2IZ4FfSXp38jlJOjhZnxgRSyLiAuApKtM6bV/LYHbh5WmfPrzde8dJ6pU0CjiZSs9wPjBN0h7Jvnsl/UWK/VgBOdiK5xvASuB+VR4yciWVnvctwKrkvW9RmZXiFSLiKSrn6OZKehC4IXnre8Ap/YMHwJnAlGRwYiUvj85eRCUYV1A5JP1Nypo/AJyW7HMFL0+J/kVJP0++xz3Ag8ACKofBNQcPgC8An5f0AH9+5HEvcDPwEJVzb8uiMir8aeBOSQ9ROR85PmX9VjCe3cPMSsc9NjMrHQebmZWOg83MSsfBZmal42Azs9JxsJlZ6TjYzKx0/j+Ea85vd5Cr5QAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "F1_score, Accuracy = metrics(fires_testy,Y_predtst, \"Logistic_reg_test\")\n",
        "print(\"Test Logistic regression F1_score=\", F1_score, \"Test Logistic regression Accuracy=\", Accuracy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "5kFQHLfPPM1C",
        "outputId": "712dba3e-eff4-4fdf-e111-4b044b994b83"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Logistic regression F1_score= 0.8260869565217391 Test Logistic regression Accuracy= 0.8666666666666667\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        ">If our p value is less than 0.05 then we reject the null hypothesis \n",
        "And if you see we are getting a p value of 0.00, that means the 'ISI' is the best feature for the calculation so we will drop the others and predict logistic regression on just ISI Feature and See what it results."
      ],
      "metadata": {
        "id": "tBKFHj73QNsH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Only ISI\n",
        "df = pd.read_csv('/content/EE559_Project-main/datasets/algerian_fires_train.csv', usecols=['ISI'])\n",
        "dftst = pd.read_csv('/content/EE559_Project-main/datasets/algerian_fires_test.csv', usecols=['ISI'])"
      ],
      "metadata": {
        "id": "ZxIJCTmJU0H-"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "plt.tight_layout()\n",
        "\n",
        "plt.figure(figsize = (10,20))\n",
        "plt.subplot(3,2,1)\n",
        "logreg = LogisticRegression()\n",
        "logreg.fit(df, fires_trainy)\n",
        "Y_pred = logreg.predict(df)\n",
        "val=accuracy_score(fires_trainy, Y_pred)\n",
        "print(\"Accuracy for train for Only ISI:\",val)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_trainy, Y_pred)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label only Train ISI');\n",
        "plt.xlabel('Predicted label only Train ISI');\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "Y_predtst = logreg.predict(dftst)\n",
        "val3=accuracy_score(fires_testy, Y_predtst)\n",
        "print(\"Accuracy for test only ISI:\",val3)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_testy, Y_predtst)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label only Test ISI');\n",
        "plt.xlabel('Predicted label only Test ISI');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 516
        },
        "id": "6I1TK0-Lg-q2",
        "outputId": "23138ecb-cc1a-4068-bbf9-2453993f5ecf"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for train for Only ISI: 0.9130434782608695\n",
            "--------------------------\n",
            "[[ 62   7]\n",
            " [  9 106]]\n",
            "Accuracy for test only ISI: 0.8666666666666667\n",
            "--------------------------\n",
            "[[33  4]\n",
            " [ 4 19]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x1440 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#1drop temperature\n",
        "fires_trainx_1 = fires_train.iloc[:,2:-1]\n",
        "fires_trainy_1 = fires_train.iloc[:,-1]\n",
        "fires_testx_1 = fires_test.iloc[:,2:-1]\n",
        "fires_testy_1 = fires_test.iloc[:,-1]\n",
        "#2drop Temp,RH\n",
        "fires_trainx_2 = fires_train.iloc[:,3:-1]\n",
        "fires_trainy_2 = fires_train.iloc[:,-1]\n",
        "fires_testx_2 = fires_test.iloc[:,3:-1]\n",
        "fires_testy_2 = fires_test.iloc[:,-1]\n",
        "\n",
        "#3drop Temp,RH, Ws\n",
        "fires_trainx_3 = fires_train.iloc[:,4:-1]\n",
        "fires_trainy_3 = fires_train.iloc[:,-1]\n",
        "fires_testx_3 = fires_test.iloc[:,4:-1]\n",
        "fires_testy_3 = fires_test.iloc[:,-1]\n"
      ],
      "metadata": {
        "id": "1aqqoTSiUEJ_"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#1drop\n",
        "###############Working \n",
        "from sklearn.metrics import accuracy_score\n",
        "plt.tight_layout()\n",
        "plt.figure(figsize = (10,20))\n",
        "plt.subplot(3,2,1)\n",
        "logreg1 = LogisticRegression()\n",
        "logreg1.fit(fires_trainx_1, fires_trainy_1)\n",
        "Y_pred1 = logreg1.predict(fires_trainx_1)\n",
        "val1=accuracy_score(fires_trainy_1, Y_pred1)\n",
        "print(\"Accuracy for train 1drop :\",val1)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_trainy_1, Y_pred1)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label 1drop');\n",
        "plt.xlabel('Predicted label 1drop');\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "Y_predtst1 = logreg1.predict(fires_testx_1)\n",
        "val12=accuracy_score(fires_testy_1,Y_predtst1)\n",
        "print(\"Accuracy for test 1drop:\",val12)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_testy_1, Y_predtst1)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label test 1drop');\n",
        "plt.xlabel('Predicted label test 1drop');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 516
        },
        "id": "qY8lpsIVWfk3",
        "outputId": "cfdf77d9-cc89-4961-e648-581c88ac0607"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for train 1drop : 0.9239130434782609\n",
            "--------------------------\n",
            "[[ 63   6]\n",
            " [  8 107]]\n",
            "Accuracy for test 1drop: 0.9\n",
            "--------------------------\n",
            "[[34  3]\n",
            " [ 3 20]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x1440 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#2drop\n",
        "plt.tight_layout()\n",
        "plt.figure(figsize = (10,20))\n",
        "plt.subplot(3,2,1)\n",
        "logreg2 = LogisticRegression()\n",
        "logreg2.fit(fires_trainx_2, fires_trainy_2)\n",
        "Y_pred2 = logreg2.predict(fires_trainx_2)\n",
        "val2=logreg2.score(fires_trainx_2,fires_trainy_2)\n",
        "print(\"Accuracy for train 2drop :\",val2)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_trainy_2, Y_pred2)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label 2drop');\n",
        "plt.xlabel('Predicted label 2drop');\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "Y_predtst2 = logreg2.predict(fires_testx_2)\n",
        "val1212=accuracy_score(fires_testy_2,Y_predtst2)\n",
        "print(\"Accuracy for test 2drop:\",val1212)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_testy_2, Y_predtst2)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label test 2drop');\n",
        "plt.xlabel('Predicted label test 2drop');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 516
        },
        "id": "cHP7TgpWdu0J",
        "outputId": "f5a843ac-a18e-455c-9fbd-4d22d0916bda"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for train 2drop : 0.9239130434782609\n",
            "--------------------------\n",
            "[[ 63   6]\n",
            " [  8 107]]\n",
            "Accuracy for test 2drop: 0.8833333333333333\n",
            "--------------------------\n",
            "[[34  3]\n",
            " [ 4 19]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x1440 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#3drop\n",
        "plt.tight_layout()\n",
        "plt.figure(figsize = (10,20))\n",
        "plt.subplot(3,2,1)\n",
        "logreg3 = LogisticRegression()\n",
        "logreg3.fit(fires_trainx_3, fires_trainy_3)\n",
        "Y_pred3 = logreg3.predict(fires_trainx_3)\n",
        "val3=logreg3.score(fires_trainx_3,fires_trainy_3)\n",
        "print(\"Accuracy for train 3drop :\",val3)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_trainy_3, Y_pred3)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label 2drop');\n",
        "plt.xlabel('Predicted label 2drop');\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "Y_predtst3 = logreg3.predict(fires_testx_3)\n",
        "val121212=accuracy_score(fires_testy_3,Y_predtst3)\n",
        "print(\"Accuracy for test 3drop:\",val121212)\n",
        "print(\"--------------------------\")\n",
        "conf_matrix = confusion_matrix(fires_testy_3, Y_predtst3)\n",
        "sns.heatmap(conf_matrix, annot=True, square=True, cmap='Blues', fmt=\"d\", cbar=True)\n",
        "print(conf_matrix)\n",
        "plt.ylabel('Actual label test 3drop');\n",
        "plt.xlabel('Predicted label test 3drop');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 516
        },
        "id": "Oj26uJOQfM7Y",
        "outputId": "9d8ce589-fc4e-4739-e9ee-061195055e85"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy for train 3drop : 0.9293478260869565\n",
            "--------------------------\n",
            "[[ 63   6]\n",
            " [  7 108]]\n",
            "Accuracy for test 3drop: 0.8666666666666667\n",
            "--------------------------\n",
            "[[33  4]\n",
            " [ 4 19]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x1440 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}