Build Linear Regression With Python

Our Example

We will build a model that predicts an exam score based on the number of hours a student studies.

Import Linear Regression

First, we import the Linear Regression model from scikit-learn.

from sklearn.linear_model import LinearRegression

We can now create a Linear Regression model in Python.

Create the Training Data

Next, we create our input data and the corresponding output values.

X = [[1], [2], [3], [4], [5]]

y = [45, 55, 65, 75, 85]

Here:

Notice that X is written as a list of lists:

[[1], [2], [3], [4], [5]]

Each inner list represents one training example and its feature.

Create the Model

Now we create a Linear Regression model object.

model = LinearRegression()

At this point, the model has been created, but it has not learned anything yet.

This distinction is important. Creating the model and training the model are two different steps.

Train the Model

Now we give the training data to the model.

model.fit(X, y)

The fit() method is where the model learns the relationship between X and y.

The model learns the coefficients needed to represent the relationship in the training data.

Make a Prediction

Our model is now trained. Suppose a new student studied for 6 hours.

prediction = model.predict([[6]])

print(prediction)

We give 6 to the trained model.

For this simple dataset, the model learns a relationship equivalent to:

Therefore, 6 hours gives a prediction of approximately 95.

The Complete Python Code

Now put all the steps together:

from sklearn.linear_model import LinearRegression

# Training data
X = [[1], [2], [3], [4], [5]]
y = [45, 55, 65, 75, 85]

# Create the model
model = LinearRegression()

# Train the model
model.fit(X, y)

# Make a prediction
prediction = model.predict([[6]])

print(prediction)

Understand the Code as a Workflow

What Does fit() Actually Do?

This is one of the most important things to understand.

model.fit(X, y)

The model looks at the training examples and learns the relationship between the input and target.

For Linear Regression, these learned parameters include the coefficients and intercept used to make predictions.

What Does predict() Do?

model.predict([[6]])

The predict() method takes new input and uses the trained model to calculate the estimated output.

Why Is X Written as [[6]]?

Beginners often get confused by this:

model.predict([[6]])

The reason is that scikit-learn expects input features in a two-dimensional structure:

Think of it as:

[
    [6]
]

The outer list represents the collection of samples. The inner list contains the features for one sample.

Understand the Four Important Lines