# Let us import some helpful libraries.
import warnings
warnings.filterwarnings("ignore")  # We will ignore warning messages to keep it clean.
import pandas as pd
import numpy as np

# Data setup
url = "https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv"
titanic = pd.read_csv(url)
print("Rows and columns:", titanic.shape)
titanic.head()

# Let us see how many people survived and who did not.
titanic['Survived'].value_counts()

# Store your age and your name
my_age = 15
my_name = "Sam"
print("My name is", my_name)
print("My age is", my_age)

# Get your favorite color from user input
favorite_color = input("What is your favorite color?")
print("Wow! I like", favorite_color, "too!")

# Show all rows where Sex is 'female'
women = titanic[titanic['Sex'] == 'female']
print("Number of women:", len(women))
women.head()

# Check for missing values
titanic.isnull().sum()

# Fill missing age data with the average age
mean_age = titanic['Age'].mean()
titanic['Age'].fillna(mean_age, inplace=True)

# Remove columns we do not need
titanic = titanic.drop(['Cabin', 'Ticket', 'Name'], axis=1)

# Change 'Sex' to a number: 0 for female, 1 for male
titanic['Sex'] = titanic['Sex'].map({'female': 0, 'male': 1})

# Simple model: Predict survival based on Sex
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression

X = titanic[['Sex']]
y = titanic['Survived']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

model = LogisticRegression()
model.fit(X_train, y_train)

# Predict on the test data and check accuracy
y_pred = model.predict(X_test)
accuracy = np.mean(y_pred == y_test)
print("Accuracy:", accuracy)

# Use more columns: Sex, Age, and Pclass
X = titanic[['Sex', 'Age', 'Pclass']]
y = titanic['Survived']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
model = LogisticRegression()
model.fit(X_train, y_train)
accuracy = model.score(X_test, y_test)
print("New accuracy:", accuracy)

# Try the model on new data with input().
sex_input = input("Enter 0 for female, 1 for male: ")
age_input = input("Enter age: ")
pclass_input = input("Enter class (1, 2, or 3): ")
# Convert inputs to numbers
user_data = np.array([[int(sex_input), float(age_input), int(pclass_input)]])
prediction = model.predict(user_data)
if prediction[0] == 1:
    print("The model predicts: Survived")
else:
    print("The model predicts: Did not survive")
    

# What if we forget random_state?
X_train2, X_test2, y_train2, y_test2 = train_test_split(X, y, test_size=0.2)
print(X_train.equals(X_train2))

# Troubleshooting: Common errors
# Try dividing by zero
try:
    result = 10 / 0
except ZeroDivisionError:
    print("Oops! You cannot divide by zero.")
    

# Your mini challenge: Try predicting survival using only 'Age' and 'Pclass'
X_challenge = titanic[['Age', 'Pclass']]
y_challenge = titanic['Survived']
X_train_c, X_test_c, y_train_c, y_test_c = train_test_split(X_challenge, y_challenge, test_size=0.2, random_state=42)
model_challenge = LogisticRegression()
model_challenge.fit(X_train_c, y_train_c)
print("Challenge accuracy:", model_challenge.score(X_test_c, y_test_c))