code

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sb
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.dummy import DummyClassifier
from sklearn import metrics
from sklearn.metrics import plot_confusion_matrix

# 1
from google.colab import files
uploaded = files.upload() 
df=pd.read_csv("Titanic.csv")
df.head()


# 2
print(df.head())
print(df.shape)
print(df.columns.values)

# target variable is whether a passenger survived
# dropped passenger number from logistic regression
# does not have relevance for survival


# 3
feature_cols = ['Class', 'Sex', 'Age']
X = df[feature_cols] # features
y = df['Survived'] # target variable
df = df.drop(columns=['Passenger'])


# 4
print(df.isnull().sum())


# 5
plt.figure(1)
sb.countplot(x='Survived', data=df)

plt.figure(2)
sb.countplot(y='Class', data=df)

plt.figure(3)
sb.countplot(x='Sex', data=df)

plt.figure(4)
sb.countplot(y='Age', data=df)

plt.show()


# 6
# Convert all categorical feature variables into dummy variables. (3)
df2 = pd.get_dummies(df, columns=['Class', 'Sex', 'Age'])
print(df2.head())


# 7
# Partition the data into train and test sets (70/30). Use random_state = 2020. (2)
X = df2.iloc[:, 1:]
y = df2.iloc[:, 0]
print(X.head())
print(y.head())

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=2020)


# 8
# Fit the training data to a logistic regression model. (2)
logReg = LogisticRegression()
logReg.fit(X_train, y_train)


# 9
# Predict the outcomes for the test set. (2)
y_pred = logReg.predict(X_test)
print(y_pred)


# 10
# Display the accuracy of your predictions for survivability. (2)
print('Accuracy:', metrics.accuracy_score(y_test, y_pred),"or 77.31%")


# 11
# Display the confusion matrix along with the labels (Yes, No). Hint: You may want to use from sklearn.metrics import plot_confusion_matrix (2)
cnf_matrix = metrics.confusion_matrix(y_test, y_pred)
print(cnf_matrix)
metrics.plot_confusion_matrix(logReg, X_test, y_test)
plt.show()

dummy_clf = DummyClassifier(strategy="most_frequent")
dummy_clf.fit(X_train, y_train)
dummy_clf.predict(X_test)
dummy_clf.score(X_test,y_test)


# 12
# Now, display the predicted value of the survivability of a male adult passenger traveling in 3rd class. (3)
#d = {'Class_1st': [0], 'Class_2nd':[0],'Class_3rd':[1], 'Class_Crew':[0],'Sex_Female': [0],'Sex_Male':[1],'Age_Adult':[1],'Age_Child':[0]}
#x_new = pd.DataFrame(data=d)
x_new = [[0,0,1,0,0,1,1,0]]
#print(x_new)
#x_new.reshape(1, -1)
predict = logReg.predict(x_new)
print(predict)


# 13
# Next, display the predicted probability of the survivability of a male adult passenger traveling in 3rd class. (3)
probability = logReg.predict_proba(x_new)
print(probability)
# The classifier predicts survival as Yes if the probability in the second column of variable ‘probability’ is greater than 0.5
# When the probability in the second column is less than 0.5, then the classifier is predicting No

print("\nThe predicted probability of survivability is: 11.14%")