{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"**Steps**\n\n*     what is classification?\n*     what is Logistic Regression?\n*     imports\n*     Handling missing values\n*     Data visualization\n*     removing outliers\n*     One Hot Encoding\n*     Feature Scaling\n*     Submission\n\n**Let us learn, how to deal with a classification problem.**\n\n\n**What is Classification?**\n\n\nWhen the target is categorical (with 2 or more classes), then we use Classification technique.\n\n\nIn Titanic dataset, the target is \"Survived\" variable with class 0 and class 1. \nclass 0 indicates \"not survived\" and class 1 indicates \"survived\".\n\n\n\n**What is Logistic Regression?**\n\n\nLogistic Regression is used for classification problems. Logistic Regression is named for the     \nfunction used at the core of the method, the Logistic function. The logistic function is also known \nas the Sigmoid function. \nIt is an S-shaped curve.","metadata":{"trusted":true}},{"cell_type":"markdown","source":"![image.png](attachment:image.png)","metadata":{},"attachments":{"image.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"check out below links for Logistic Regression in depth.\n\nhttps://youtu.be/L_xBe7MbPwk\n\nhttps://youtu.be/uFfsSgQgerw\n","metadata":{}},{"cell_type":"markdown","source":"Import the libraries and modules.","metadata":{}},{"cell_type":"code","source":"import numpy as np \nimport pandas as pd\nimport matplotlib.pyplot as plt # data visualization\nimport seaborn as sns # data visualization\nfrom sklearn.preprocessing import StandardScaler, MinMaxScaler, OneHotEncoder\n\nimport warnings\nwarnings.filterwarnings('ignore')\n\nplt.style.use('ggplot')","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.870892Z","iopub.execute_input":"2022-07-08T17:14:35.871364Z","iopub.status.idle":"2022-07-08T17:14:35.883016Z","shell.execute_reply.started":"2022-07-08T17:14:35.871326Z","shell.execute_reply":"2022-07-08T17:14:35.881921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Now import train data and test data.","metadata":{}},{"cell_type":"code","source":"train = pd.read_csv('../input/titanic/train.csv')\ntest  = pd.read_csv('../input/titanic/test.csv')","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.890896Z","iopub.execute_input":"2022-07-08T17:14:35.891449Z","iopub.status.idle":"2022-07-08T17:14:35.914710Z","shell.execute_reply.started":"2022-07-08T17:14:35.891410Z","shell.execute_reply":"2022-07-08T17:14:35.913695Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train.Survived.value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.916298Z","iopub.execute_input":"2022-07-08T17:14:35.916845Z","iopub.status.idle":"2022-07-08T17:14:35.927019Z","shell.execute_reply.started":"2022-07-08T17:14:35.916801Z","shell.execute_reply":"2022-07-08T17:14:35.925773Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"To know the information of the dataset, we use info() function. It gives information about variables, number of non-null count, datatype of each columns and memory usage.","metadata":{}},{"cell_type":"code","source":"print(train.info())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.932343Z","iopub.execute_input":"2022-07-08T17:14:35.933305Z","iopub.status.idle":"2022-07-08T17:14:35.950507Z","shell.execute_reply.started":"2022-07-08T17:14:35.933249Z","shell.execute_reply":"2022-07-08T17:14:35.949265Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(test.info())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.955827Z","iopub.execute_input":"2022-07-08T17:14:35.956506Z","iopub.status.idle":"2022-07-08T17:14:35.973437Z","shell.execute_reply.started":"2022-07-08T17:14:35.956452Z","shell.execute_reply":"2022-07-08T17:14:35.971758Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"describe() function gives statistical details of the dataset.","metadata":{}},{"cell_type":"code","source":"print(train.describe())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:35.975511Z","iopub.execute_input":"2022-07-08T17:14:35.976029Z","iopub.status.idle":"2022-07-08T17:14:36.010998Z","shell.execute_reply.started":"2022-07-08T17:14:35.975991Z","shell.execute_reply":"2022-07-08T17:14:36.009436Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(test.describe())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.013210Z","iopub.execute_input":"2022-07-08T17:14:36.013822Z","iopub.status.idle":"2022-07-08T17:14:36.044394Z","shell.execute_reply.started":"2022-07-08T17:14:36.013776Z","shell.execute_reply":"2022-07-08T17:14:36.043225Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Lets check the null values in the dataset.","metadata":{}},{"cell_type":"code","source":"print(train.isnull().sum())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.047550Z","iopub.execute_input":"2022-07-08T17:14:36.048073Z","iopub.status.idle":"2022-07-08T17:14:36.057598Z","shell.execute_reply.started":"2022-07-08T17:14:36.048021Z","shell.execute_reply":"2022-07-08T17:14:36.056470Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(test.isnull().sum())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.059252Z","iopub.execute_input":"2022-07-08T17:14:36.059808Z","iopub.status.idle":"2022-07-08T17:14:36.072926Z","shell.execute_reply.started":"2022-07-08T17:14:36.059767Z","shell.execute_reply":"2022-07-08T17:14:36.071908Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are null values in train dataset and test dataset.\n","metadata":{"trusted":true}},{"cell_type":"code","source":"print(train.columns)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.075287Z","iopub.execute_input":"2022-07-08T17:14:36.075802Z","iopub.status.idle":"2022-07-08T17:14:36.081501Z","shell.execute_reply.started":"2022-07-08T17:14:36.075764Z","shell.execute_reply":"2022-07-08T17:14:36.080688Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(test.columns)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.119425Z","iopub.execute_input":"2022-07-08T17:14:36.120171Z","iopub.status.idle":"2022-07-08T17:14:36.127015Z","shell.execute_reply.started":"2022-07-08T17:14:36.120108Z","shell.execute_reply":"2022-07-08T17:14:36.125914Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Lets remove the variables which are not significant in this data.","metadata":{}},{"cell_type":"code","source":"train.drop(columns=['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)\ntest.drop(columns= ['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace= True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.136091Z","iopub.execute_input":"2022-07-08T17:14:36.136642Z","iopub.status.idle":"2022-07-08T17:14:36.146909Z","shell.execute_reply.started":"2022-07-08T17:14:36.136590Z","shell.execute_reply":"2022-07-08T17:14:36.145492Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Fill the null values with mean, median or mode according to the requirements.","metadata":{}},{"cell_type":"code","source":"train['Age'].median()\n","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.156259Z","iopub.execute_input":"2022-07-08T17:14:36.156733Z","iopub.status.idle":"2022-07-08T17:14:36.164804Z","shell.execute_reply.started":"2022-07-08T17:14:36.156688Z","shell.execute_reply":"2022-07-08T17:14:36.163606Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Embarked'].mode()[0]","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.180152Z","iopub.execute_input":"2022-07-08T17:14:36.180806Z","iopub.status.idle":"2022-07-08T17:14:36.189255Z","shell.execute_reply.started":"2022-07-08T17:14:36.180748Z","shell.execute_reply":"2022-07-08T17:14:36.188128Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Age'].fillna(train['Age'].median(), inplace=True)\ntrain['Embarked'].fillna(train['Embarked'].mode()[0], inplace=True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.197670Z","iopub.execute_input":"2022-07-08T17:14:36.198052Z","iopub.status.idle":"2022-07-08T17:14:36.207544Z","shell.execute_reply.started":"2022-07-08T17:14:36.198020Z","shell.execute_reply":"2022-07-08T17:14:36.206139Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(train.isnull().sum())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.218032Z","iopub.execute_input":"2022-07-08T17:14:36.218669Z","iopub.status.idle":"2022-07-08T17:14:36.229861Z","shell.execute_reply.started":"2022-07-08T17:14:36.218584Z","shell.execute_reply":"2022-07-08T17:14:36.227559Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are no null values in train dataset\n","metadata":{}},{"cell_type":"code","source":"test['Age'].median()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.235751Z","iopub.execute_input":"2022-07-08T17:14:36.236355Z","iopub.status.idle":"2022-07-08T17:14:36.247328Z","shell.execute_reply.started":"2022-07-08T17:14:36.236301Z","shell.execute_reply":"2022-07-08T17:14:36.245249Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Fare'].median()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.255116Z","iopub.execute_input":"2022-07-08T17:14:36.255796Z","iopub.status.idle":"2022-07-08T17:14:36.265097Z","shell.execute_reply.started":"2022-07-08T17:14:36.255743Z","shell.execute_reply":"2022-07-08T17:14:36.263792Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Age'].fillna(test['Age'].median(), inplace=True)\ntest['Fare'].fillna(test['Fare'].median(), inplace=True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.271131Z","iopub.execute_input":"2022-07-08T17:14:36.271742Z","iopub.status.idle":"2022-07-08T17:14:36.280677Z","shell.execute_reply.started":"2022-07-08T17:14:36.271687Z","shell.execute_reply":"2022-07-08T17:14:36.279782Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(test.isnull().sum())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.290952Z","iopub.execute_input":"2022-07-08T17:14:36.293012Z","iopub.status.idle":"2022-07-08T17:14:36.305310Z","shell.execute_reply.started":"2022-07-08T17:14:36.292918Z","shell.execute_reply":"2022-07-08T17:14:36.303339Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" There are no null values in test dataset","metadata":{}},{"cell_type":"markdown","source":"value_counts() gives count of unique values.","metadata":{}},{"cell_type":"code","source":"train['Survived'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.314424Z","iopub.execute_input":"2022-07-08T17:14:36.315251Z","iopub.status.idle":"2022-07-08T17:14:36.326887Z","shell.execute_reply.started":"2022-07-08T17:14:36.315200Z","shell.execute_reply":"2022-07-08T17:14:36.324990Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Pclass'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.329784Z","iopub.execute_input":"2022-07-08T17:14:36.330609Z","iopub.status.idle":"2022-07-08T17:14:36.344512Z","shell.execute_reply.started":"2022-07-08T17:14:36.330560Z","shell.execute_reply":"2022-07-08T17:14:36.343132Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Sex'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.347181Z","iopub.execute_input":"2022-07-08T17:14:36.347671Z","iopub.status.idle":"2022-07-08T17:14:36.358749Z","shell.execute_reply.started":"2022-07-08T17:14:36.347590Z","shell.execute_reply":"2022-07-08T17:14:36.357585Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['SibSp'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.364884Z","iopub.execute_input":"2022-07-08T17:14:36.365469Z","iopub.status.idle":"2022-07-08T17:14:36.376487Z","shell.execute_reply.started":"2022-07-08T17:14:36.365414Z","shell.execute_reply":"2022-07-08T17:14:36.375307Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Parch'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.383597Z","iopub.execute_input":"2022-07-08T17:14:36.384278Z","iopub.status.idle":"2022-07-08T17:14:36.395081Z","shell.execute_reply.started":"2022-07-08T17:14:36.384227Z","shell.execute_reply":"2022-07-08T17:14:36.393482Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train['Embarked'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.402838Z","iopub.execute_input":"2022-07-08T17:14:36.403304Z","iopub.status.idle":"2022-07-08T17:14:36.415822Z","shell.execute_reply.started":"2022-07-08T17:14:36.403255Z","shell.execute_reply":"2022-07-08T17:14:36.414780Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Pclass'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.421044Z","iopub.execute_input":"2022-07-08T17:14:36.421577Z","iopub.status.idle":"2022-07-08T17:14:36.433187Z","shell.execute_reply.started":"2022-07-08T17:14:36.421540Z","shell.execute_reply":"2022-07-08T17:14:36.432108Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Sex'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.442764Z","iopub.execute_input":"2022-07-08T17:14:36.443589Z","iopub.status.idle":"2022-07-08T17:14:36.453114Z","shell.execute_reply.started":"2022-07-08T17:14:36.443548Z","shell.execute_reply":"2022-07-08T17:14:36.451990Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['SibSp'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.460339Z","iopub.execute_input":"2022-07-08T17:14:36.460784Z","iopub.status.idle":"2022-07-08T17:14:36.471343Z","shell.execute_reply.started":"2022-07-08T17:14:36.460749Z","shell.execute_reply":"2022-07-08T17:14:36.469877Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Parch'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.479817Z","iopub.execute_input":"2022-07-08T17:14:36.480452Z","iopub.status.idle":"2022-07-08T17:14:36.490385Z","shell.execute_reply.started":"2022-07-08T17:14:36.480415Z","shell.execute_reply":"2022-07-08T17:14:36.489188Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test['Embarked'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.496661Z","iopub.execute_input":"2022-07-08T17:14:36.497327Z","iopub.status.idle":"2022-07-08T17:14:36.506833Z","shell.execute_reply.started":"2022-07-08T17:14:36.497287Z","shell.execute_reply":"2022-07-08T17:14:36.505848Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Lets do Data Visualization for train dataset and test dataset","metadata":{"trusted":true}},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.countplot(x='Survived', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.516276Z","iopub.execute_input":"2022-07-08T17:14:36.516895Z","iopub.status.idle":"2022-07-08T17:14:36.658080Z","shell.execute_reply.started":"2022-07-08T17:14:36.516854Z","shell.execute_reply":"2022-07-08T17:14:36.656812Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"From the above figure, we can observe that survived people are less compared to not-survived people.","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.countplot(x='Sex', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.661324Z","iopub.execute_input":"2022-07-08T17:14:36.661770Z","iopub.status.idle":"2022-07-08T17:14:36.804380Z","shell.execute_reply.started":"2022-07-08T17:14:36.661733Z","shell.execute_reply":"2022-07-08T17:14:36.803397Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"males are more compared to females","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.countplot(x='Survived', hue='Sex', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.805681Z","iopub.execute_input":"2022-07-08T17:14:36.806130Z","iopub.status.idle":"2022-07-08T17:14:36.971335Z","shell.execute_reply.started":"2022-07-08T17:14:36.806098Z","shell.execute_reply":"2022-07-08T17:14:36.970272Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.countplot(x='Survived', hue='Pclass', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:36.972674Z","iopub.execute_input":"2022-07-08T17:14:36.973163Z","iopub.status.idle":"2022-07-08T17:14:37.165303Z","shell.execute_reply.started":"2022-07-08T17:14:36.973128Z","shell.execute_reply":"2022-07-08T17:14:37.163800Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.boxplot(x='Survived', y= 'Age', hue='Sex', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:37.169249Z","iopub.execute_input":"2022-07-08T17:14:37.169668Z","iopub.status.idle":"2022-07-08T17:14:37.425661Z","shell.execute_reply.started":"2022-07-08T17:14:37.169598Z","shell.execute_reply":"2022-07-08T17:14:37.423702Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.boxplot(x='Pclass', y= 'Fare', data= train)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:37.430912Z","iopub.execute_input":"2022-07-08T17:14:37.431316Z","iopub.status.idle":"2022-07-08T17:14:37.620123Z","shell.execute_reply.started":"2022-07-08T17:14:37.431283Z","shell.execute_reply":"2022-07-08T17:14:37.618357Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Fare is high in Pclass 1 compared to Pclass 2 and Pclass 3","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.countplot(x='Sex', data= test)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:37.622233Z","iopub.execute_input":"2022-07-08T17:14:37.622754Z","iopub.status.idle":"2022-07-08T17:14:37.768318Z","shell.execute_reply.started":"2022-07-08T17:14:37.622704Z","shell.execute_reply":"2022-07-08T17:14:37.766543Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\nsns.boxplot(x='Pclass', y= 'Fare', data= test)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:37.770328Z","iopub.execute_input":"2022-07-08T17:14:37.770758Z","iopub.status.idle":"2022-07-08T17:14:37.954110Z","shell.execute_reply.started":"2022-07-08T17:14:37.770720Z","shell.execute_reply":"2022-07-08T17:14:37.952811Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Check outliers in the data","metadata":{}},{"cell_type":"code","source":"train.plot(kind='box', figsize= (10,8))","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:37.956173Z","iopub.execute_input":"2022-07-08T17:14:37.956704Z","iopub.status.idle":"2022-07-08T17:14:38.203835Z","shell.execute_reply.started":"2022-07-08T17:14:37.956649Z","shell.execute_reply":"2022-07-08T17:14:38.201770Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are outliers in Age, SibSp, Parch and Fare variables. I am using clip() funtion to remove the outliers. Parch variable is removed because more than 75% of the values are 0.","metadata":{}},{"cell_type":"code","source":"cols= ['Age', 'SibSp', 'Parch', 'Fare']\n\ntrain[cols]= train[cols].clip(lower= train[cols].quantile(0.15), upper= train[cols].quantile(0.85), axis=1)\n\ntrain.drop(columns=['Parch'], axis=1, inplace=True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.206078Z","iopub.execute_input":"2022-07-08T17:14:38.206538Z","iopub.status.idle":"2022-07-08T17:14:38.236142Z","shell.execute_reply.started":"2022-07-08T17:14:38.206489Z","shell.execute_reply":"2022-07-08T17:14:38.235069Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train.plot(kind='box', figsize= (10,8)) ","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.237766Z","iopub.execute_input":"2022-07-08T17:14:38.238166Z","iopub.status.idle":"2022-07-08T17:14:38.476815Z","shell.execute_reply.started":"2022-07-08T17:14:38.238128Z","shell.execute_reply":"2022-07-08T17:14:38.475722Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are no outliers in the train dataset","metadata":{}},{"cell_type":"code","source":"test.plot(kind='box', figsize= (10,8))","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.478755Z","iopub.execute_input":"2022-07-08T17:14:38.479122Z","iopub.status.idle":"2022-07-08T17:14:38.702065Z","shell.execute_reply.started":"2022-07-08T17:14:38.479085Z","shell.execute_reply":"2022-07-08T17:14:38.700829Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are outliers in Age, SibSp, Parch and Fare variables. clip() funtion is used to remove the outliers. Parch variable is removed because more than 75% of the values are 0.","metadata":{}},{"cell_type":"code","source":"test[cols]= test[cols].clip(lower= test[cols].quantile(0.15), upper= test[cols].quantile(0.85), axis=1)\n\ntest.drop(columns=['Parch'], axis=1, inplace=True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.704208Z","iopub.execute_input":"2022-07-08T17:14:38.704653Z","iopub.status.idle":"2022-07-08T17:14:38.732401Z","shell.execute_reply.started":"2022-07-08T17:14:38.704594Z","shell.execute_reply":"2022-07-08T17:14:38.730759Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test.plot(kind='box', figsize= (10,8))  ","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.733857Z","iopub.execute_input":"2022-07-08T17:14:38.734204Z","iopub.status.idle":"2022-07-08T17:14:38.940256Z","shell.execute_reply.started":"2022-07-08T17:14:38.734171Z","shell.execute_reply":"2022-07-08T17:14:38.938698Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"There are no outliers in the test dataset","metadata":{}},{"cell_type":"markdown","source":"One Hot Encoding is used to convert categorical data into numerical data.\n","metadata":{}},{"cell_type":"markdown","source":"![image.png](attachment:88c88f92-eb81-4b29-84f2-bfd0423ef063.png)","metadata":{},"attachments":{"88c88f92-eb81-4b29-84f2-bfd0423ef063.png":{"image/png":"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"}}},{"cell_type":"code","source":"cat_features = ['Pclass', 'Sex', 'Embarked']","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.942918Z","iopub.execute_input":"2022-07-08T17:14:38.943506Z","iopub.status.idle":"2022-07-08T17:14:38.948768Z","shell.execute_reply.started":"2022-07-08T17:14:38.943447Z","shell.execute_reply":"2022-07-08T17:14:38.947139Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train= pd.get_dummies(train, columns=cat_features, drop_first= True)\ntest= pd.get_dummies(test, columns=cat_features, drop_first= True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.950998Z","iopub.execute_input":"2022-07-08T17:14:38.951368Z","iopub.status.idle":"2022-07-08T17:14:38.974847Z","shell.execute_reply.started":"2022-07-08T17:14:38.951331Z","shell.execute_reply":"2022-07-08T17:14:38.973814Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train.head()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.976172Z","iopub.execute_input":"2022-07-08T17:14:38.976683Z","iopub.status.idle":"2022-07-08T17:14:38.993746Z","shell.execute_reply.started":"2022-07-08T17:14:38.976645Z","shell.execute_reply":"2022-07-08T17:14:38.992288Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test.head()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:38.995337Z","iopub.execute_input":"2022-07-08T17:14:38.995702Z","iopub.status.idle":"2022-07-08T17:14:39.013829Z","shell.execute_reply.started":"2022-07-08T17:14:38.995661Z","shell.execute_reply":"2022-07-08T17:14:39.012344Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Now, lets split the data.","metadata":{}},{"cell_type":"code","source":"X_train= train.iloc[:, 1:]\ny_train= train['Survived'].values.reshape(-1,1)\n\nX_test= test","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:39.015253Z","iopub.execute_input":"2022-07-08T17:14:39.016237Z","iopub.status.idle":"2022-07-08T17:14:39.025530Z","shell.execute_reply.started":"2022-07-08T17:14:39.016177Z","shell.execute_reply":"2022-07-08T17:14:39.023910Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Feature Scaling is used to standardize the independent variables present in the data in a fixed range.","metadata":{}},{"cell_type":"code","source":"ss = StandardScaler()\n# ss = MinMaxScaler(feature_range=(0,1))","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:39.027564Z","iopub.execute_input":"2022-07-08T17:14:39.028057Z","iopub.status.idle":"2022-07-08T17:14:39.034721Z","shell.execute_reply.started":"2022-07-08T17:14:39.028015Z","shell.execute_reply":"2022-07-08T17:14:39.033498Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"features= ['Age', 'SibSp', 'Fare']\n\nX_train[features] = ss.fit_transform(X_train[features])\nX_test[features] = ss.transform(X_test[features])","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:39.036149Z","iopub.execute_input":"2022-07-08T17:14:39.036655Z","iopub.status.idle":"2022-07-08T17:14:39.064327Z","shell.execute_reply.started":"2022-07-08T17:14:39.036591Z","shell.execute_reply":"2022-07-08T17:14:39.063263Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"!pip install minisom","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:39.065864Z","iopub.execute_input":"2022-07-08T17:14:39.066431Z","iopub.status.idle":"2022-07-08T17:14:47.349014Z","shell.execute_reply.started":"2022-07-08T17:14:39.066395Z","shell.execute_reply":"2022-07-08T17:14:47.347659Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pylab\nfrom minisom import MiniSom","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:47.351156Z","iopub.execute_input":"2022-07-08T17:14:47.351532Z","iopub.status.idle":"2022-07-08T17:14:47.357253Z","shell.execute_reply.started":"2022-07-08T17:14:47.351489Z","shell.execute_reply":"2022-07-08T17:14:47.355745Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:47.359091Z","iopub.execute_input":"2022-07-08T17:14:47.359536Z","iopub.status.idle":"2022-07-08T17:14:47.392542Z","shell.execute_reply.started":"2022-07-08T17:14:47.359501Z","shell.execute_reply":"2022-07-08T17:14:47.391286Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"minisom = MiniSom(50,50,X_train.values.shape[1])\nminisom.random_weights_init(X_train.values)\nminisom.train_random(X_train.values,10000,verbose=True)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:47.396291Z","iopub.execute_input":"2022-07-08T17:14:47.396854Z","iopub.status.idle":"2022-07-08T17:14:55.225223Z","shell.execute_reply.started":"2022-07-08T17:14:47.396803Z","shell.execute_reply":"2022-07-08T17:14:55.223847Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pylab.figure(figsize=(15,13))\npylab.bone()\n\npylab.pcolor(minisom.distance_map().T)\npylab.colorbar()\n\n# Red died and Green survived and Blue is unknown (need to predict)\nmarkers = ['x', 'o', '']\ncolors = ['r', 'g', '']\n\nfor i, x in enumerate(X_train.values):\n     w = minisom.winner(x)\n     pylab.plot(w[0]+0.5,w[1]+0.5, markers[int(y_train[i][0])], markersize=5, markeredgecolor = colors[int(y_train[i][0])], markerfacecolor = 'None', markeredgewidth=2)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:55.227134Z","iopub.execute_input":"2022-07-08T17:14:55.227594Z","iopub.status.idle":"2022-07-08T17:14:57.680455Z","shell.execute_reply.started":"2022-07-08T17:14:55.227545Z","shell.execute_reply":"2022-07-08T17:14:57.679241Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\nX_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.33, random_state=42)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:57.682180Z","iopub.execute_input":"2022-07-08T17:14:57.682547Z","iopub.status.idle":"2022-07-08T17:14:57.691150Z","shell.execute_reply.started":"2022-07-08T17:14:57.682513Z","shell.execute_reply":"2022-07-08T17:14:57.690217Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.linear_model import LogisticRegression\n\nlr = LogisticRegression()\nlr.fit(X_train, y_train.ravel())\nlr_pred = lr.predict_proba(X_val)[:,1] ","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:57.693713Z","iopub.execute_input":"2022-07-08T17:14:57.694415Z","iopub.status.idle":"2022-07-08T17:14:57.728257Z","shell.execute_reply.started":"2022-07-08T17:14:57.694372Z","shell.execute_reply":"2022-07-08T17:14:57.727099Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(lr.score(X_val, y_val))","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:57.730863Z","iopub.execute_input":"2022-07-08T17:14:57.731789Z","iopub.status.idle":"2022-07-08T17:14:57.747653Z","shell.execute_reply.started":"2022-07-08T17:14:57.731731Z","shell.execute_reply":"2022-07-08T17:14:57.743736Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Decision Tree Model\n\n![image.png](attachment:e67b4d66-8d4c-438a-82de-2d9ade2c2679.png)","metadata":{},"attachments":{"e67b4d66-8d4c-438a-82de-2d9ade2c2679.png":{"image/png":"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"}}},{"cell_type":"code","source":"from sklearn.ensemble import RandomForestClassifier\n\nrf = RandomForestClassifier()\nrf.fit(X_train, y_train.ravel())\nrf_pred = rf.predict_proba(X_val)[:,1] ","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:57.749943Z","iopub.execute_input":"2022-07-08T17:14:57.751826Z","iopub.status.idle":"2022-07-08T17:14:58.024481Z","shell.execute_reply.started":"2022-07-08T17:14:57.751781Z","shell.execute_reply":"2022-07-08T17:14:58.023574Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(rf.score(X_val, y_val))","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.026343Z","iopub.execute_input":"2022-07-08T17:14:58.026845Z","iopub.status.idle":"2022-07-08T17:14:58.053520Z","shell.execute_reply.started":"2022-07-08T17:14:58.026795Z","shell.execute_reply":"2022-07-08T17:14:58.052587Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.058455Z","iopub.execute_input":"2022-07-08T17:14:58.059000Z","iopub.status.idle":"2022-07-08T17:14:58.065194Z","shell.execute_reply.started":"2022-07-08T17:14:58.058965Z","shell.execute_reply":"2022-07-08T17:14:58.063119Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"(lr_pred + rf_pred) / 2","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.067311Z","iopub.execute_input":"2022-07-08T17:14:58.067909Z","iopub.status.idle":"2022-07-08T17:14:58.085521Z","shell.execute_reply.started":"2022-07-08T17:14:58.067870Z","shell.execute_reply":"2022-07-08T17:14:58.084499Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ensemble_pred = [1 if i > 0.5 else 0 for i in ((lr_pred + rf_pred) / 2)]","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.087778Z","iopub.execute_input":"2022-07-08T17:14:58.088146Z","iopub.status.idle":"2022-07-08T17:14:58.094582Z","shell.execute_reply.started":"2022-07-08T17:14:58.088098Z","shell.execute_reply":"2022-07-08T17:14:58.092979Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"accuracy_score(ensemble_pred, y_val.ravel())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.096666Z","iopub.execute_input":"2022-07-08T17:14:58.097426Z","iopub.status.idle":"2022-07-08T17:14:58.110973Z","shell.execute_reply.started":"2022-07-08T17:14:58.097371Z","shell.execute_reply":"2022-07-08T17:14:58.109683Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"confusion_matrix(ensemble_pred, y_val.ravel())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.113466Z","iopub.execute_input":"2022-07-08T17:14:58.114178Z","iopub.status.idle":"2022-07-08T17:14:58.125980Z","shell.execute_reply.started":"2022-07-08T17:14:58.114125Z","shell.execute_reply":"2022-07-08T17:14:58.124823Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"precision_score(ensemble_pred, y_val.ravel())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.128100Z","iopub.execute_input":"2022-07-08T17:14:58.128499Z","iopub.status.idle":"2022-07-08T17:14:58.140253Z","shell.execute_reply.started":"2022-07-08T17:14:58.128450Z","shell.execute_reply":"2022-07-08T17:14:58.138377Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"recall_score(ensemble_pred, y_val.ravel())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.142114Z","iopub.execute_input":"2022-07-08T17:14:58.142857Z","iopub.status.idle":"2022-07-08T17:14:58.153242Z","shell.execute_reply.started":"2022-07-08T17:14:58.142802Z","shell.execute_reply":"2022-07-08T17:14:58.151689Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"f1_score(ensemble_pred, y_val.ravel())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.154941Z","iopub.execute_input":"2022-07-08T17:14:58.155462Z","iopub.status.idle":"2022-07-08T17:14:58.167575Z","shell.execute_reply.started":"2022-07-08T17:14:58.155426Z","shell.execute_reply":"2022-07-08T17:14:58.165802Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"A confusion matrix is a tool designed to help us understand a little better how well our classifier is performing. An *accuracy score*, like that returned by kaggle for our submission file, lets us know a number indicating what ratio of predictions were correct (0 is not one classification was correct, and 1 is perfect!). The confusion matrix does the same thing, but goes into a little more detail; this time it provides us with four values:\n* The number of times our classifier produced **true negatives** (TN) the model correctly predicts the negatives class\n* The number of times our classifier produced **true positives** (TP) the model correctly predicts the positive class\n* The number of times our classifier produced **false positives** (FP), a type I error the model incorrectly predicts the positive class\n* The number of times our classifier produced **false negatives** (FN), a type II error the model incorrectly predicts the negatives class\n\nwhich scikit-learn returns in the following format, hence the name matrix (note that there is no standard convention for arrangement of this matrix):\n\n![image.png](attachment:d7636ca6-9e5b-4a08-9ab0-bd6c01850e21.png)\n\n![image.png](attachment:8fec5ff1-9ec8-46a8-880b-fd9dec11e68f.png)\n\nThe *accuracy* is given by $\\frac{(TN + TP)}{(TN + TP + FP +FN)}$, in other words, the true values divided by all the values. And finally, another measure one may come across is the **$F_1$ score**, which is given by:\n\n$$ F_1 = 2\\frac{precision . recall}{precision + recall}$$\n\n\nwhere the *precision* is given by $\\frac{TP}{TP + FP}$, and *recall* by $\\frac{TP}{TP + FN}$.\n\nThese Wikipedia pages have excellent descriptions of the meaning of these terms: \n* [Confusion matrix](https://en.wikipedia.org/wiki/Confusion_matrix)\n* [False positives and false negatives](https://en.wikipedia.org/wiki/False_positives_and_false_negatives)\n* [Type I and type II errors](https://en.wikipedia.org/wiki/Type_I_and_type_II_errors)\n* [Receiver operating characteristic](https://en.wikipedia.org/wiki/Receiver_operating_characteristic)\n* [F1 score](https://en.wikipedia.org/wiki/F1_score)\n","metadata":{},"attachments":{"d7636ca6-9e5b-4a08-9ab0-bd6c01850e21.png":{"image/png":"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"},"8fec5ff1-9ec8-46a8-880b-fd9dec11e68f.png":{"image/png":"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"}}},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_pred_lr = lr.predict_proba(X_test)[:,1]\ny_pred_rf = rf.predict_proba(X_test)[:,1] \n\npredictions = (y_pred_lr + y_pred_rf) / 2\npredictions = [1 if i > 0.5 else 0 for i in predictions]","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.169260Z","iopub.execute_input":"2022-07-08T17:14:58.169901Z","iopub.status.idle":"2022-07-08T17:14:58.204204Z","shell.execute_reply.started":"2022-07-08T17:14:58.169853Z","shell.execute_reply":"2022-07-08T17:14:58.202850Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test = pd.read_csv('../input/titanic/test.csv')","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.205956Z","iopub.execute_input":"2022-07-08T17:14:58.206324Z","iopub.status.idle":"2022-07-08T17:14:58.220942Z","shell.execute_reply.started":"2022-07-08T17:14:58.206290Z","shell.execute_reply":"2022-07-08T17:14:58.219816Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission= pd.DataFrame({'PassengerId' : test['PassengerId'], 'Survived': predictions })\n\nprint(submission.head())","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.223025Z","iopub.execute_input":"2022-07-08T17:14:58.223539Z","iopub.status.idle":"2022-07-08T17:14:58.235930Z","shell.execute_reply.started":"2022-07-08T17:14:58.223503Z","shell.execute_reply":"2022-07-08T17:14:58.234506Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.Survived.value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.238246Z","iopub.execute_input":"2022-07-08T17:14:58.238757Z","iopub.status.idle":"2022-07-08T17:14:58.254548Z","shell.execute_reply.started":"2022-07-08T17:14:58.238701Z","shell.execute_reply":"2022-07-08T17:14:58.253240Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"filename= 'titanic predictions.csv'\nsubmission.to_csv(filename, index=False)","metadata":{"execution":{"iopub.status.busy":"2022-07-08T17:14:58.255892Z","iopub.execute_input":"2022-07-08T17:14:58.256239Z","iopub.status.idle":"2022-07-08T17:14:58.268177Z","shell.execute_reply.started":"2022-07-08T17:14:58.256206Z","shell.execute_reply":"2022-07-08T17:14:58.266866Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" **If you enjoy this notebook, please share and give an upvote. Any suggestions or comments are appreciated. Happy Learning :)**","metadata":{"trusted":true}}]}