# First, let us be sure warnings do not distract us.
import warnings
warnings.filterwarnings("ignore")

print("Ready to learn!")

# Let us try some simple math in Python. 
x = 10
y = 3 * x + 7
print("When x =", x, "then y =", y)

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

# Let us look at the column names. 
df.columns

# Let us see basic statistics. 
df.describe()

# What are we predicting? The target: MedHouseVal.
df['MedHouseVal'].head()

# Let us try simple linear regression with just one feature.
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression

feature = 'MedInc'  # Median income
X = df[[feature]]
y = df['MedHouseVal']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

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

print("Model is trained!")

# Let us predict home values for test data.
y_pred = model.predict(X_test)
print("First 5 Predictions:", y_pred[:5])
print("First 5 Actual Values:", list(y_test[:5]))

# Calculate how well our model did.
from sklearn.metrics import mean_squared_error
mse = mean_squared_error(y_test, y_pred)
print("Mean Squared Error:", mse)

# What does our line look like? Let us plot it.
import matplotlib.pyplot as plt

plt.scatter(X_test, y_test, color="blue", label="Actual")
plt.plot(X_test, y_pred, color="red", label="Prediction")
plt.xlabel("Median Income")
plt.ylabel("Median House Value")
plt.title("Home Value Prediction - Simple Regression")
plt.legend()
plt.show()

# Let us pick a few features.
features = ['MedInc', 'AveRooms', 'AveOccup']
X = df[features]

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

multi_model = LinearRegression()
multi_model.fit(X_train, y_train)

multi_pred = multi_model.predict(X_test)

print("First 5 Predictions:", multi_pred[:5])

# Let us measure this model's performance.
multi_mse = mean_squared_error(y_test, multi_pred)
print("Multi-feature MSE:", multi_mse)
print("Root MSE:", multi_mse ** 0.5)

# View feature importance (the weights).
importance = dict(zip(features, multi_model.coef_))
print("Feature Weights:")
for feat, coef in importance.items():
    print(feat, ":", coef)
    

# Predict for your own values!
example = [[5.0, 6.0, 3.0]]
prediction = multi_model.predict(example)
print("Predicted home value:", prediction[0])

# Try your own prediction using input().
income = float(input("Enter median income: "))
rooms = float(input("Enter average rooms: "))
occup = float(input("Enter average occupancy: "))
user_pred = multi_model.predict([[income, rooms, occup]])
print("Predicted house value:", user_pred[0])

# Troubleshooting: What if you get a funny error?
try:
    broken = multi_model.predict([[None, 6.0, 3.0]])
    print(broken)
except Exception as e:
    print("Oops! Prediction failed. Reason:", e)