Add files via upload

Author: Algorithmica

2019-october/7.visual eda/bivariate-eda.py

@@ -0,0 +1,25 @@
+import pandas as pd
+import os
+import seaborn as sns
+
+dir = 'E:/'
+titanic_train = pd.read_csv(os.path.join(dir, 'train.csv'))
+print(titanic_train.info())
+print(titanic_train.columns)
+
+#explore bivariate relationships: categorical vs categorical 
+sns.factorplot(x="Sex", hue="Survived", data=titanic_train, kind="count", size=6)
+#sns.factorplot(x="Survived", hue="Sex", data=titanic_train, kind="count", size=6)
+sns.factorplot(x="Pclass", hue="Survived", data=titanic_train, kind="count", size=6)
+sns.factorplot(x="Embarked", hue="Survived", data=titanic_train, kind="count", size=6)
+
+#explore bivariate relationships: continuous vs categorical
+sns.FacetGrid(titanic_train, hue="Survived",size=8).map(sns.kdeplot, "Fare").add_legend()
+sns.FacetGrid(titanic_train, col="Survived",size=8).map(sns.kdeplot, "Fare").add_legend()
+sns.FacetGrid(titanic_train, hue="Survived",size=8).map(sns.kdeplot, "Age").add_legend()
+
+#explore bivariate relationships: continuous vs continuous 
+sns.jointplot(x="Age", y="Fare", data=titanic_train)
+
+features = ['Sex', 'Pclass', 'Embarked', 'SibSp', 'Parch', 'Fare','Age', 'Survived']
+sns.pairplot(titanic_train[features])

2019-october/7.visual eda/multivariate-eda.py

@@ -0,0 +1,31 @@
+import pandas as pd
+import os
+import seaborn as sns
+
+dir = 'E:/'
+titanic_train = pd.read_csv(os.path.join(dir, 'train.csv'))
+print(titanic_train.info())
+print(titanic_train.columns)
+
+#understand grid cell formation
+g = sns.FacetGrid(titanic_train, col="Sex") 
+g.map(sns.kdeplot, "Age")
+g.map(sns.boxplot, "Age")
+g = sns.FacetGrid(titanic_train, row="Sex") 
+g.map(sns.kdeplot, "Fare")
+g = sns.FacetGrid(titanic_train, hue="Survived")
+g.map(sns.kdeplot, "Fare").add_legend()
+
+#is age have an impact on survived for each sex group?
+g = sns.FacetGrid(titanic_train, col="Sex", hue="Survived") 
+g.map(sns.kdeplot, "Age").add_legend()
+
+#is age have an impact on survived for each pclass and sex group?
+g = sns.FacetGrid(titanic_train, row="Pclass", col="Sex", hue="Survived") 
+g.map(sns.kdeplot, "Age").add_legend()
+
+g = sns.FacetGrid(titanic_train, row="Pclass", col="Sex", hue="Survived") 
+g.map(sns.kdeplot, "Age")
+
+tmp = titanic_train[['Fare','Age','Parch','SibSp']]
+sns.heatmap(tmp.corr())

2019-october/7.visual eda/univariate-eda.py

@@ -0,0 +1,27 @@
+import pandas as pd
+import os
+import seaborn as sns
+
+dir = 'E:/'
+titanic_train = pd.read_csv(os.path.join(dir, 'train.csv'))
+print(titanic_train.info())
+print(titanic_train.columns)
+
+#categorical columns: numerical EDA
+pd.crosstab(index=titanic_train["Survived"], columns="count")
+
+#categorical columns: visual EDA
+sns.countplot(x='Survived',data=titanic_train)
+sns.countplot(x='Pclass',data=titanic_train)
+sns.countplot(x='Sex',data=titanic_train)
+
+#continuous features: visual EDA
+titanic_train['Fare'].describe()
+sns.boxplot(x='Fare',data=titanic_train)
+sns.distplot(titanic_train['Fare'])
+sns.distplot(titanic_train['Fare'], hist=False)
+sns.distplot(titanic_train['Age'], hist=False)
+sns.boxplot(x='Age',data=titanic_train)
+
+sns.distplot(titanic_train['SibSp'], hist=False)
+sns.boxplot(x='SibSp',data=titanic_train)