# Always suppress warnings for smoother runs
import warnings; warnings.filterwarnings("ignore")
import numpy as np
np.random.seed(42)

# Data setup
import pandas as pd
url = 'https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv'
df = pd.read_csv(url)
print(df.shape)
print(df.head(3))

# Quick data info: see types and missing values
df.info()

# Count missing values in each column
df.isnull().sum().sort_values(ascending=False)

# Remove rows where Price or Manufacturer is missing
df_clean = df.dropna(subset=['Price', 'Manufacturer'])
print('Rows after drop:', df_clean.shape[0])

# Fill remaining missing numeric values with their column median
num_cols = df_clean.select_dtypes(include='number').columns
df_clean[num_cols] = df_clean[num_cols].fillna(df_clean[num_cols].median())

# Basic statistics for price
df_clean['Price'].describe()

# Plot the price distribution
import seaborn as sns
import matplotlib.pyplot as plt
sns.histplot(df_clean['Price'], bins=20, kde=True)
plt.title('Distribution of Car Prices')
plt.xlabel('Price in $1000s')
plt.show()

# Average price by manufacturer (top 10)
mean_price = df_clean.groupby('Manufacturer')['Price'].mean().sort_values(ascending=False).head(10)
mean_price.plot(kind='bar', figsize=(8,4), color='skyblue')
plt.ylabel('Average Price ($1000s)')
plt.title('Top 10 Manufacturers by Average Car Price')
plt.show()

# Scatter plot: price vs. horsepower
sns.scatterplot(data=df_clean, x='Horsepower', y='Price', alpha=0.7)
plt.title('Car Price vs Horsepower')
plt.xlabel('Horsepower')
plt.ylabel('Price ($1000s)')
plt.show()

# Boxplot: price by car type
sns.boxplot(data=df_clean, x='Type', y='Price')
plt.title('Price by Car Type')
plt.xlabel('Type')
plt.ylabel('Price ($1000s)')
plt.xticks(rotation=30)
plt.show()

# Correlation matrix: how features relate to price and each other
corr = df_clean.select_dtypes(include='number').corr()
sns.heatmap(corr, annot=True, fmt='.1f', cmap='Blues', vmin=-1, vmax=1)
plt.title('Correlation Heatmap (numeric features)')
plt.show()

# Prepare data for modeling: select features and convert categories
features = ['Horsepower', 'RPM', 'EngineSize', 'MPG.city', 'MPG.highway', 'Weight', 'Length', 'Type']
df_model = df_clean[features + ['Price']].dropna()
df_model = pd.get_dummies(df_model, columns=['Type'], drop_first=True)
print(df_model.columns)

# Split into training and test sets
from sklearn.model_selection import train_test_split
X = df_model.drop('Price', axis=1)
y = df_model['Price']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
print('Training samples:', X_train.shape[0])

# Linear regression model: predict price from features
from sklearn.linear_model import LinearRegression
model = LinearRegression()
model.fit(X_train, y_train)
print('Model coefficients:', model.coef_)
print('Intercept:', model.intercept_)

# Predict and compare to actual prices on test data
y_pred = model.predict(X_test)
results = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})
print(results.head())

# Evaluate accuracy with Root Mean Squared Error
from sklearn.metrics import mean_squared_error
import numpy as np
rmse = np.sqrt(mean_squared_error(y_test, y_pred))
print('Root Mean Squared Error:', round(rmse, 2))

# Plot actual vs. predicted prices
plt.scatter(y_test, y_pred, alpha=0.7)
plt.xlabel('Actual Price ($1000s)')
plt.ylabel('Predicted Price ($1000s)')
plt.title('Actual vs Predicted Car Prices')
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'r--')
plt.show()