In this tutorial, you’ll learn how to implement binary classification with Keras using the Sequential model. This beginner-friendly guide walks you through data prep, model building, training, and prediction—all using standalone Keras.
Binary classification is a supervised learning task where we classify data into one of two classes (e.g., Spam or Not Spam, Fraud or Legitimate).
Binary classification is a type of machine learning task where the goal is to classify inputs into one of two categories.
Real-World Examples:
-
Spam vs Not Spam (email filtering)
-
Disease vs No Disease (medical diagnosis)
-
Customer will churn vs Will stay (business analytics)
-
Fraudulent vs Legitimate transaction (finance)
How It Works:
A binary classifier learns from labeled data, where each example is tagged as class 0 or 1. It then tries to predict the correct label for new, unseen examples.
Output:
Typically:
-
The model outputs a probability between 0 and 1
-
Then applies a threshold (e.g., 0.5) to decide:
-
> 0.5→ Class 1 -
≤ 0.5→ Class 0
-
This tutorial demonstrates how to build a simple binary classifier using only standalone Keras.
Why Use Keras for Binary Classification?
Keras is a powerful and beginner-friendly deep learning library that makes building neural networks super simple. It’s built for ease of use, speed of experimentation, and scalability.
Top Reasons to Use Keras:
1. Easy to Learn & Use
-
Intuitive API — you can build a model in just a few lines of code.
-
Ideal for beginners and prototyping.
2. Built on Top of TensorFlow
-
Combines simplicity with the power and performance of TensorFlow.
-
Easily scalable for large datasets or GPU training.
3. Modular & Flexible
-
You can plug in different layers, loss functions, optimizers, etc., like Lego blocks.
-
Supports both Sequential and Functional APIs.
4. Well-Documented & Actively Maintained
-
Tons of tutorials, guides, and community support.
-
Maintained by the TensorFlow team at Google.
5. Real-World Ready
-
Used in industry and research for everything from image classification to NLP and reinforcement learning.
Model architecture:
New to Keras?
Explore the official docs here: Keras Documentation
Building a Binary Classification Model with Keras:
1. Install Dependencies:
pip install keras
2. Import Required Libraries:
import numpy as np from keras.models import Sequential from keras.layers import Dense from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler
3. Generate Sample Dataset:
# Generate synthetic data np.random.seed(42) X = np.random.rand(1000, 2) # 1000 samples, 2 features y = (X[:, 0] + X[:, 1] > 1).astype(int) # Label: 1 if sum > 1, else 0 # Split into train & test sets X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) # Normalize the data scaler = StandardScaler() X_train = scaler.fit_transform(X_train) X_test = scaler.transform(X_test)
4. Build the Keras Model:
# Define the model
model = Sequential([
Dense(16, activation='relu', input_shape=(2,)), # Input layer
Dense(8, activation='relu'), # Hidden layer
Dense(1, activation='sigmoid') # Output layer for binary classification
])
# Compile the model
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
5. Train the Model:
# Train the model model.fit(X_train, y_train, epochs=20, batch_size=32, validation_data=(X_test, y_test))
Output:
6. Evaluate the Model:
loss, accuracy = model.evaluate(X_test, y_test)
print(f"Test Accuracy: {accuracy:.4f}")
output:
7/7 ━━━━━━━━━━━━━━━━━━━━ 0s 14ms/step - accuracy: 0.9892 - loss: 0.0632 Test Accuracy: 0.9950
7. Make Predictions:
# Predict on new data
new_data = np.array([[0.2, 0.8]]) # Example input
new_data = scaler.transform(new_data) # Scale it
prediction = model.predict(new_data)
predicted_class = (prediction > 0.5).astype(int)
print(f"Predicted Class: {predicted_class[0][0]}")
output:
1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 144ms/step Predicted Class: 1