# Let us start with a simple data list.
ages = [17, 21, 34, 42, 15, 29]

# Print the data.
print('Original ages data:', ages)

# Feature engineering often means creating new features.
# For example, let us find out who is an adult.

is_adult = []

for age in ages:
    if age >= 18:
        is_adult.append(True)
    else:
        is_adult.append(False)

print('Is each person an adult?', is_adult)

# Let us add two lists together to make a dataset.
names = ['Anna', 'Ben', 'Cleo', 'Dan', 'Elle', 'Felix']

for name, age, adult in zip(names, ages, is_adult):
    print(name, age, adult)
    

# Let us try engineering a feature from text.

favorite_fruits = ['apple', 'banana', 'apple', 'orange', 'banana', 'apple']
# Let us count how many times 'apple' appears.
apple_count = favorite_fruits.count('apple')

print('Number of times apple appears:', apple_count)

# Handling missing or empty data is important.
# Let us see how to fill in missing ages.

ages_with_missing = [17, 21, None, 42, None, 29]

filled_ages = []
for age in ages_with_missing:
    if age is None:
        filled_ages.append(18)  # We use eighteen as a default age.
    else:
        filled_ages.append(age)

print('Filled ages:', filled_ages)

# Categorical features are often easier to use as numbers.
# Let us turn fruits into numbers.

fruit_to_number = {'apple': 0, 'banana': 1, 'orange': 2}
fruit_numbers = []
for fruit in favorite_fruits:
    fruit_numbers.append(fruit_to_number[fruit])

print('Fruit as numbers:', fruit_numbers)

# Creating features from math or logic is common.
# Let us find the difference from the oldest age.

oldest = max(filled_ages)
age_gaps = [oldest - age for age in filled_ages]

print('Difference from oldest:', age_gaps)

# Scaling numbers helps machine learning.
min_age = min(filled_ages)
max_age = max(filled_ages)

normalized_ages = []
for age in filled_ages:
    normalized = (age - min_age) / (max_age - min_age)
    normalized_ages.append(normalized)

print('Normalized ages:', normalized_ages)

# Sometimes we want features to show a relationship.
siblings = [1, 2, 0, 3, 1, 2]

# Let us make a new feature: age per sibling.
age_per_sibling = []
for age, sib in zip(filled_ages, siblings):
    if sib == 0:
        age_per_sibling.append(age)
    else:
        age_per_sibling.append(age // sib)

print('Age per sibling:', age_per_sibling)

# Feature engineering often means removing outliers.
all_ages = [17, 21, 201, 42, 18, 29]  # 201 is probably a data error.
clean_ages = []

for age in all_ages:
    if age < 120:
        clean_ages.append(age)

print('Ages with outliers removed:', clean_ages)

# Sometimes features are combined to make even better clues.
# Let us combine age and siblings into one string.
combined_feat = []
for age, sib in zip(filled_ages, siblings):
    result = str(age) + '_sib_' + str(sib)
    combined_feat.append(result)

print('Combined features:', combined_feat)

# Feature selection means keeping only helpful features.
features = {'age': filled_ages, 'adult': is_adult, 'siblings': siblings}

selected_features = {}
for key in features:
    if key != 'siblings':
        selected_features[key] = features[key]

print('Selected features:', selected_features)

# Let us ask the user for a new feature value.
city = input('Which city do you live in? ')

print('City entered:', city)

# Practice: Can you fill in your favorite animal and make a new feature?
animal = input('What is your favorite animal? ')
fav_animals = [animal] * len(names)

print('Everyone loves:', fav_animals)

# Mini Project Part 1: Build a tiny dataset.
student_data = [
    {'name': 'Anna', 'age': 17, 'score': 88, 'city': 'Boston'},
    {'name': 'Ben', 'age': 21, 'score': 92, 'city': 'Dallas'},
    {'name': 'Cleo', 'age': 34, 'score': 76, 'city': 'Boston'},
    {'name': 'Dan', 'age': 42, 'score': 81, 'city': 'Austin'},
]

for student in student_data:
    print(student)
    

# Mini Project Part 2: Engineer new features for our students.

for student in student_data:
    student['is_adult'] = student['age'] >= 18
    student['passed'] = student['score'] >= 80

# Print with new features
for student in student_data:
    print(student)
    

# Let us filter our student data for adults who passed.
adults_passed = []
for student in student_data:
    if student['is_adult'] and student['passed']:
        adults_passed.append(student['name'])

print('Adults who passed:', adults_passed)

# Sorting is simple but powerful.
sorted_students = sorted(student_data, key=lambda s: s['score'], reverse=True)

for student in sorted_students:
    print(student['name'], student['score'])
    

# Final tip: See what features are available.
print('All features for first student:', list(student_data[0].keys()))