# Let us start by making sure warnings will not interrupt us!
import warnings
warnings.filterwarnings('ignore')

# Comments start with # and are ignored by Python.
# Use comments to explain your code.
print("Hello, California!")

# Let us assign some values.
location = "California"
median_price = 584000
is_expensive = True
print(location, median_price, is_expensive)

# You can do basic math with Python.
a = 300000
b = 450000
total = a + b
average = total / 2
print("Total:", total)
print("Average:", average)

# Creating and printing a list of values
prices = [330000, 439000, 503200, 580000]
print(prices)

# Access the first value with index 0
print(prices[0])

# Lists can hold text too
cities = ["San Diego", "Los Angeles", "San Jose"]
print(cities)

# Try printing the last city
print(cities[-1])

# Let us handle an error: What if we use a wrong index?
try:
    print(cities[5])
except IndexError:
    print("That index does not exist!")
    

# Data setup: Load California housing data
from sklearn.datasets import fetch_california_housing
import pandas as pd
cal = fetch_california_housing(as_frame=True)
df = cal.frame
print("Shape:", df.shape)
df.head()

# What columns does the data have?
print(df.columns.tolist())

# Check for missing values (empty spaces)
print(df.isnull().sum())

# Simple statistics: What is the average house value?
mean_value = df['MedHouseVal'].mean()
print("Average house value:", mean_value)

# Selecting data: homes worth over $3,000,000
expensive = df[df['MedHouseVal'] > 3.0]
print("Number of expensive areas:", len(expensive))
expensive.head()

# Add a new column: value divided by number of rooms
df['Value_per_room'] = df['MedHouseVal'] / df['AveRooms']
df[['MedHouseVal', 'AveRooms', 'Value_per_room']].head()

# Remove the new column to clean up
del df['Value_per_room']
print('Value_per_room' in df.columns)

# Loop through values: first 5 house values
values = df['MedHouseVal'].head()
for val in values:
    print(val)
    

# List comprehension: double each value
nums = [1, 2, 3]
doubles = [n * 2 for n in nums]
print(doubles)

# Sort by median house value (lowest to highest)
sorted_df = df.sort_values('MedHouseVal')
sorted_df[['MedHouseVal']].head()

# Combine values with zip: city and value
sample_cities = ["A", "B", "C"]
sample_values = [0.5, 1.1, 2.3]
for c, v in zip(sample_cities, sample_values):
    print(c, v)
    

# Use input to predict: How many rooms? (simple version)
rooms = int(input("Enter average rooms in house: "))
prediction = rooms * 65000
print("Predicted value:", prediction)

# Mini-project, part 1: Prepare training and testing data
from sklearn.model_selection import train_test_split
X = df.drop('MedHouseVal', axis=1)
y = df['MedHouseVal']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
print(X_train.shape, X_test.shape)

# Mini-project, part 2: Build and test a prediction model
from sklearn.linear_model import LinearRegression
model = LinearRegression()
model.fit(X_train, y_train)
preds = model.predict(X_test)
print("First 5 predictions:", preds[:5])

# How good is our model? Calculate error
from sklearn.metrics import mean_squared_error
error = mean_squared_error(y_test, preds, squared=False)
print("Root Mean Squared Error:", error)

# Some troubleshooting: What if our columns do not match?
try:
    wrong_X = df.drop('MedHouseVal', axis=1).drop('AveRooms', axis=1)
    model.predict(wrong_X.head())
except Exception as e:
    print("Oops! Columns in the model and new data must match.")
    

# Tip: Use describe() to see quick stats
df.describe()

# Challenge: Predict value for a made-up house
my_data = {
    'MedInc': [6],
    'HouseAge': [30],
    'AveRooms': [8],
    'AveBedrms': [1],
    'Population': [100],
    'AveOccup': [2],
    'Latitude': [34],
    'Longitude': [-118]
}
df_input = pd.DataFrame(my_data)
print(model.predict(df_input))