Variational Quantum Classifier
A 2-qubit classifier trained with the parameter-shift rule. Watch the decision boundary form as training progresses — this is a live computation, not an animation.
Dataset
Layers
0.50
80
Epoch0 / 80
Loss1.4446
Accuracy0.4000
Epoch 0; loss 1.4446; accuracy 40.0 percent.
Learning mode
Mission briefingAdvanced / 10 min
Teach a quantum model
A variational classifier is a small adjustable lens: training turns the lens until classes separate.
Train for a few epochs, pause, then step manually to see how loss and boundary co-evolve.
Definitions on demand
Loss
A scalar penalty measuring disagreement between model predictions and labels.
Learning rate
The multiplier controlling each parameter update's size.
Decision boundary
The set of inputs where the model is equally uncertain between classes.
Decision Boundary
P(class 1)
0.01.0
Loss Curve
Train to see the loss curve.