From 7aae872089a46edeafd8f25de8683beb39bca48a Mon Sep 17 00:00:00 2001 From: Shubham Gautam <57455438+shubhamgautam1211@users.noreply.github.com> Date: Fri, 1 Oct 2021 01:14:03 +0530 Subject: [PATCH] Python debugging file added for project --- debugging_eclipse_details.ipynb | 739 ++++++++++++++++++++++++++++++++ 1 file changed, 739 insertions(+) create mode 100644 debugging_eclipse_details.ipynb diff --git a/debugging_eclipse_details.ipynb b/debugging_eclipse_details.ipynb new file mode 100644 index 0000000..31d583d --- /dev/null +++ b/debugging_eclipse_details.ipynb @@ -0,0 +1,739 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "aml_lab4_titanic.ipynb", + "provenance": [], + "collapsed_sections": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + } + }, + "cells": [ + { + "cell_type": "code", + "metadata": { + "id": "iIlPfcZaEaes" + }, + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn import preprocessing\n", + "from sklearn.model_selection import cross_validate\n", + "from sklearn.preprocessing import StandardScaler\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.metrics import accuracy_score\n", + "from sklearn.model_selection import cross_val_score\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn import svm\n", + "from statistics import mean\n", + "from sklearn.model_selection import cross_val_predict\n", + "from sklearn.metrics import confusion_matrix\n", + "from sklearn.metrics import precision_score, recall_score\n", + "from sklearn.metrics import f1_score\n", + "from sklearn.metrics import precision_recall_curve\n", + "from sklearn.metrics import accuracy_score" + ], + "execution_count": 22, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "uw5z3ZY59IVs" + }, + "source": [ + "df = pd.read_csv('Titanic dataset.csv')" + ], + "execution_count": 23, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ZUCfCQvn9Jje", + "outputId": "b98d24cd-62c0-48a9-e472-5f7665d91b9b" + }, + "source": [ + "df.nunique()" + ], + "execution_count": 24, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "pclass 3\n", + "survived 2\n", + "name 1307\n", + "sex 2\n", + "age 98\n", + "sibsp 7\n", + "parch 8\n", + "ticket 929\n", + "fare 281\n", + "cabin 186\n", + "embarked 3\n", + "dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 24 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "GtizlIcM9Ktm" + }, + "source": [ + "df['sex'] = df['sex'].replace([\"female\", \"male\"], [0, 1])\n", + "df['embarked'] = df['embarked'].replace(['S', 'C', 'Q'], [1, 2, 3])" + ], + "execution_count": 25, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 453 + }, + "id": "dZPjFDs89MCr", + "outputId": "2020fea9-bcbf-4389-f138-4eada95d041d" + }, + "source": [ + "df.hist(figsize=(6,6))\n", + "plt.tight_layout(pad=0.5)\n", + "plt.show()" + ], + "execution_count": 26, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 369 + }, + "id": "sk_MX46F9PL-", + "outputId": "dce17b9f-ce75-49d5-d73d-7c1c4ac8a77c" + }, + "source": [ + "corr = df.corr() \n", + "plt.figure(figsize=(6,5))\n", + "sns.heatmap(corr,annot=True)\n", + "plt.tight_layout()\n", + "plt.show()" + ], + "execution_count": 27, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "Oac5PrGG9QXq", + "outputId": "1f5997a6-27ca-466c-c4ab-9a0d048f7065" + }, + "source": [ + "df.drop(['pclass','cabin','name','ticket'],axis=1,inplace=True)\n", + "df.head(5)" + ], + "execution_count": 28, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
survivedsexagesibspparchfareembarked
01029.000000211.33751.0
1110.916712151.55001.0
2002.000012151.55001.0
30130.000012151.55001.0
40025.000012151.55001.0
\n", + "
" + ], + "text/plain": [ + " survived sex age sibsp parch fare embarked\n", + "0 1 0 29.0000 0 0 211.3375 1.0\n", + "1 1 1 0.9167 1 2 151.5500 1.0\n", + "2 0 0 2.0000 1 2 151.5500 1.0\n", + "3 0 1 30.0000 1 2 151.5500 1.0\n", + "4 0 0 25.0000 1 2 151.5500 1.0" + ] + }, + "metadata": {}, + "execution_count": 28 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "yqjMSUBp9Tdh", + "outputId": "e90b5364-9ce2-464f-802a-aeae56b80f19" + }, + "source": [ + "imp = SimpleImputer(missing_values=np.nan,strategy='mean')\n", + "df2 = df.copy()\n", + "df2 = imp.fit_transform(df2)\n", + "df2 = pd.DataFrame(df2,columns=df.columns)\n", + "\n", + "df2.head()" + ], + "execution_count": 29, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
survivedsexagesibspparchfareembarked
01.00.029.00000.00.0211.33751.0
11.01.00.91671.02.0151.55001.0
20.00.02.00001.02.0151.55001.0
30.01.030.00001.02.0151.55001.0
40.00.025.00001.02.0151.55001.0
\n", + "
" + ], + "text/plain": [ + " survived sex age sibsp parch fare embarked\n", + "0 1.0 0.0 29.0000 0.0 0.0 211.3375 1.0\n", + "1 1.0 1.0 0.9167 1.0 2.0 151.5500 1.0\n", + "2 0.0 0.0 2.0000 1.0 2.0 151.5500 1.0\n", + "3 0.0 1.0 30.0000 1.0 2.0 151.5500 1.0\n", + "4 0.0 0.0 25.0000 1.0 2.0 151.5500 1.0" + ] + }, + "metadata": {}, + "execution_count": 29 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "jhuzfu_Z9T2E", + "outputId": "454cab39-e60a-49f6-a6ac-50dad75c7735" + }, + "source": [ + "df2.isnull().sum()" + ], + "execution_count": 30, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "survived 0\n", + "sex 0\n", + "age 0\n", + "sibsp 0\n", + "parch 0\n", + "fare 0\n", + "embarked 0\n", + "dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 30 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "0gntJwil9Vi5" + }, + "source": [ + "def outlier_treatment(datacolumn):\n", + " sorted(datacolumn)\n", + " Q1,Q3 = np.percentile(datacolumn , [25,75])\n", + " IQR = Q3 - Q1\n", + " lower_range = Q1 - (1.5 * IQR)\n", + " upper_range = Q3 + (1.5 * IQR)\n", + " return lower_range,upper_range" + ], + "execution_count": 31, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "7C31ZIgY9ZpJ", + "outputId": "0b9efea4-f936-4521-8fb8-5cc066ebd590" + }, + "source": [ + "outlier_cols = [\"fare\"] \n", + "for col in outlier_cols:\n", + " l,u = outlier_treatment(df2[col])\n", + " df3 = df2[ (df2[col] > u) | (df2[col] < l) ]\n", + " df2.drop(df3.index , inplace=True)\n", + "\n", + "df2.shape" + ], + "execution_count": 32, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(1138, 7)" + ] + }, + "metadata": {}, + "execution_count": 32 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "knDVU05V9azi", + "outputId": "cdea6438-0d2b-4f5e-afa0-5717b9997d7e" + }, + "source": [ + "X_feature = df2.loc[:, df2.columns != 'survived']\n", + "Y_target = df2.loc[:, df2.columns == 'survived']\n", + "\n", + "print(X_feature.shape)\n", + "print(Y_target.shape)" + ], + "execution_count": 33, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "(1138, 6)\n", + "(1138, 1)\n" + ] + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Z8wZGrqW9b0D", + "outputId": "6629781f-b1b8-4430-cd9f-473d3c806418" + }, + "source": [ + "from sklearn.model_selection import train_test_split\n", + "X_train, X_test, Y_train, Y_test = train_test_split(X_feature, Y_target, test_size = 0.2,random_state=12)\n", + "print(X_train.shape)\n", + "print(X_test.shape)\n", + "print(Y_train.shape)\n", + "print(Y_test.shape)" + ], + "execution_count": 34, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "(910, 6)\n", + "(228, 6)\n", + "(910, 1)\n", + "(228, 1)\n" + ] + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "h2QmxNrf9dBN" + }, + "source": [ + "from sklearn.naive_bayes import GaussianNB \n", + "classifier = GaussianNB() " + ], + "execution_count": 35, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "BNayL-7G9hJf", + "outputId": "1f6ee953-5924-4735-ce39-bc9c7f894313" + }, + "source": [ + "Train_pred = classifier.fit(X_train, Y_train.values.ravel())\n", + "Train_pred" + ], + "execution_count": 36, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "GaussianNB(priors=None, var_smoothing=1e-09)" + ] + }, + "metadata": {}, + "execution_count": 36 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "zPsaSIU69huV", + "outputId": "4ec2474f-eac5-4bcf-e080-15289a2b0305" + }, + "source": [ + "Test_pred = classifier.fit(X_test, Y_test.values.ravel())\n", + "Test_pred" + ], + "execution_count": 37, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "GaussianNB(priors=None, var_smoothing=1e-09)" + ] + }, + "metadata": {}, + "execution_count": 37 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "9x3xvXf29kzr", + "outputId": "1bdc4cc9-278e-42d8-ff8b-2ab802dbb0ac" + }, + "source": [ + "prediction_train_gauss = round(Train_pred.score(X_train, Y_train) * 100, 2)\n", + "\n", + "print(\"Accuracy Score after Gaussian Naive Bayes (Training subset) : \",prediction_train_gauss)\n", + "\n", + "prediction_test_gauss = round(Test_pred.score(X_test, Y_test) * 100, 2)\n", + "\n", + "print(\"Accuracy Score after Gaussian Naive Bayes (Testing subset) : \",prediction_test_gauss)" + ], + "execution_count": 38, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy Score after Gaussian Naive Bayes (Training subset) : 76.92\n", + "Accuracy Score after Gaussian Naive Bayes (Testing subset) : 76.75\n" + ] + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "wr4yXcXr9lQV", + "outputId": "8f1fbf87-2f62-426b-96fe-fdb5aaed5ea1" + }, + "source": [ + "print('Probability of each class') \n", + "print('Survive = 0: %.2f' % classifier.class_prior_[0])\n", + "print('Survive = 1: %.2f' % classifier.class_prior_[1])" + ], + "execution_count": 39, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Probability of each class\n", + "Survive = 0: 0.62\n", + "Survive = 1: 0.38\n" + ] + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "oQz84oiZ9qZP", + "outputId": "fa66ce3e-b54e-4ca2-f42a-30bafdeb3da9" + }, + "source": [ + "y_pred = classifier.predict(X_test)\n", + "cf_matrix = confusion_matrix(Y_test, y_pred)\n", + "cf_matrix" + ], + "execution_count": 40, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([[118, 23],\n", + " [ 30, 57]])" + ] + }, + "metadata": {}, + "execution_count": 40 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "43kP9_PE9uvj", + "outputId": "2fd9dcfc-3381-47c1-d3f4-9e7bfc8be049" + }, + "source": [ + "from sklearn.metrics import precision_score\n", + "from sklearn.metrics import recall_score\n", + "from sklearn.metrics import f1_score\n", + "\n", + "ps = precision_score(Y_test, y_pred)\n", + "print(\"Precision Score : \",ps)\n", + "\n", + "rs = recall_score(Y_test, y_pred)\n", + "print(\"Recall Score : \",rs)\n", + "\n", + "fs = f1_score(Y_test, y_pred)\n", + "print(\"F1 Score \",fs)" + ], + "execution_count": 41, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Precision Score : 0.7125\n", + "Recall Score : 0.6551724137931034\n", + "F1 Score 0.6826347305389221\n" + ] + } + ] + } + ] +} \ No newline at end of file