What is Regression in Machine Learning in Python with sklearn

In machine learning, regression is a method to predict continuous values based on input data. Using Python's sklearn library, you can create models that learn patterns from data to estimate numbers like prices or temperatures.

How It Works

Regression in machine learning is like trying to draw a smooth line or curve through points on a graph to guess missing values. Imagine you want to predict the price of a house based on its size. You look at many houses with known sizes and prices, then find a pattern that connects size to price.

Once the pattern is found, you can use it to predict the price of a new house just by knowing its size. The model learns from past examples and tries to minimize the difference between its guesses and the actual prices.

Example

This example shows how to use LinearRegression from sklearn to predict house prices based on size.

python

from sklearn.linear_model import LinearRegression
import numpy as np

# Sample data: house sizes (in square feet) and prices (in $1000s)
sizes = np.array([500, 750, 1000, 1250, 1500]).reshape(-1, 1)
prices = np.array([150, 200, 250, 300, 350])

# Create and train the model
model = LinearRegression()
model.fit(sizes, prices)

# Predict price for a new house size
new_size = np.array([[1200]])
predicted_price = model.predict(new_size)
print(f"Predicted price for 1200 sq ft: ${predicted_price[0]*1000:.2f}")

Output

Predicted price for 1200 sq ft: $280000.00

When to Use

Use regression when you want to predict a number, not a category. It works well for tasks like estimating house prices, forecasting sales, predicting temperatures, or any case where the output is a continuous value.

For example, a car dealer might use regression to predict car prices based on mileage and age. A farmer could predict crop yield based on rainfall and temperature.

Key Points

• Regression predicts continuous numeric values from input data.
• It finds a relationship between features and the target number.
• sklearn provides easy-to-use regression models like LinearRegression.
• Useful for price prediction, forecasting, and any numeric estimation.

Key Takeaways

Regression predicts continuous values by learning patterns from data.

Use sklearn's LinearRegression to build simple regression models in Python.

Regression is ideal for tasks like price prediction and forecasting.

The model minimizes errors between predicted and actual values.

Regression outputs numbers, not categories.