Variational Quantum Classifier with PennyLane

Variational Quantum Classifier with PennyLane

This PennyLane tutorial builds a variational quantum classifier, a parameterized quantum circuit trained to label data, and is one of the foundational demonstrations of quantum machine learning. The variational classifier works like a quantum analogue of a neural network: classical data is encoded into a circuit through a feature map, a trainable ansatz of parameterized gates transforms the state, and a measurement produces an output that is interpreted as a class label. The tutorial trains such a model on two classic problems, learning the parity function and classifying the Iris dataset, showing the full supervised-learning loop: encoding inputs, defining the model circuit and a bias term, choosing a cost function, and optimizing the parameters with gradient descent using PennyLane's automatic differentiation. It walks through batching the training data, monitoring accuracy on a validation set, and interpreting the learned decision boundary. By grounding quantum machine learning in concrete classification tasks with a clear train/evaluate workflow, the tutorial gives a reusable template for building and training quantum classifiers in PennyLane and is a natural next step after the basic optimization tutorials.

Run it

pip install -r requirements.txt
python demo.py

Source and license

Imported from demonstrations_v2/tutorial_variational_classifier/demo.py in PennyLaneAI/demos at c52c0abeb5122218aa96b38eea848864cce7323f, under the Apache License 2.0. Original authors: Xanadu and the PennyLane community. The upstream LICENSE is included alongside this example.