Understand the Python Code

The Complete Code

First, look at the complete program before we break it into individual parts.

from sklearn.linear_model import LogisticRegression

# Training data
X = [
    [1],
    [2],
    [3],
    [5],
    [6],
    [7]
]

# Target labels
y = [
    0,
    0,
    0,
    1,
    1,
    1
]

# Create the model
model = LogisticRegression()

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

# New data
new_student = [[6]]

# Make a prediction
prediction = model.predict(new_student)

# Get probabilities
probability = model.predict_proba(new_student)

print("Classes:", model.classes_)
print("Prediction:", prediction)
print("Probabilities:", probability)

Line 1 — Import Logistic Regression

from sklearn.linear_model import LogisticRegression

This line imports the LogisticRegression class from scikit-learn.

We need this because scikit-learn provides the Machine Learning algorithm for us.

The X Variable — Input Features

X = [
    [1],
    [2],
    [3],
    [5],
    [6],
    [7]
]

X contains the input features that the model will use to learn.

In our example, X represents the number of hours each student studied.

Notice that every value is inside another pair of brackets:

[1]
[2]
[3]

This is because scikit-learn expects X to be a two-dimensional collection:

The y Variable — Target Labels

y = [
    0,
    0,
    0,
    1,
    1,
    1
]

y contains the correct answers associated with the training data.

So the training examples can be understood as:

Create the Model

model = LogisticRegression()

This creates a Logistic Regression model object and stores it inside the variable called model.

Think of it as creating an empty learner. It exists, but it has not learned from our data yet.

Train the Model With fit()

model.fit(X, y)

This is one of the most important lines in the entire program.

fit() tells the model to learn from the training examples.

Give the Model New Data

new_student = [[6]]

Now we create a new input that the model has to classify.

In this example:

The important point is that we are not giving the correct answer to the model.

We only provide the input:

[[6]]

The model must decide whether this belongs to Class 0 or Class 1.

Make the Prediction

prediction = model.predict(new_student)

The predict() method asks the trained model to choose the most likely class for the new data.

Because we defined Class 1 as Pass:

Get the Prediction Probabilities

probability = model.predict_proba(new_student)

Instead of asking only which class the model chooses, we can ask for the probability of every class.

For example:

[[0.10 0.90]]

So the model predicts Class 1 because its probability is higher.

Understand classes_

print("Classes:", model.classes_)

The classes_ attribute tells us the order of the classes used by the model.

For our example:

Classes: [0 1]

Therefore, if the probability output is:

[[0.10 0.90]]

we know:

The print() Statements

print("Classes:", model.classes_)
print("Prediction:", prediction)
print("Probabilities:", probability)

These lines simply display the results in the terminal.

For example, the terminal might show:

Classes: [0 1]
Prediction: [1]
Probabilities: [[0.10 0.90]]

Read the Output Like a Human

Let's translate the output into normal language.

And:

The Whole Program in Simple Words

One Important Thing to Remember

The model does not memorize the answer for the new student.

We trained it using examples such as:

Then we gave it a new input:

The model uses the pattern it learned from the training data to make a prediction.

Check Your Understanding