coding

Author: mhjensen

doc/src/week13/programs/qsvmq.py

@@ -0,0 +1,62 @@
+# Custom Kernel Function: The custom_kernel function uses the Radial
+# Basis Function (RBF) kernel to mimic the behavior of a quantum
+# feature map. In a real QSVM, this would be replaced by a quantum
+# kernel computed from a quantum circuit.
+
+# Training and Evaluation: The SVM is trained using the custom kernel,
+# and its performance is evaluated on the test set.  This approach
+# approximates the behavior of a QSVM using classical methods. It does
+# not provide the potential computational advantages that a true QSVM
+# might offer when run on quantum hardware.
+
+from sklearn import datasets
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+from sklearn.svm import SVC
+from sklearn.metrics import accuracy_score
+
+# Load dataset
+iris = datasets.load_iris()
+X, y = iris.data, iris.target
+
+# For simplicity, select only two classes
+X, y = X[y != 2], y[y != 2]
+
+# Standardize features
+scaler = StandardScaler()
+X = scaler.fit_transform(X)
+
+# Split into training and testing sets
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+# Train classical SVM
+clf = SVC(kernel='linear')
+clf.fit(X_train, y_train)
+
+# Predict and evaluate
+y_pred = clf.predict(X_test)
+accuracy = accuracy_score(y_test, y_pred)
+print(f'Classical SVM Accuracy: {accuracy * 100:.2f}%')
+
+# Quantum kernels map input data into a higher-dimensional space using
+# quantum operations. While we can’t perform true quantum operations
+# without a quantum simulator or hardware, we can simulate the effect
+# of a quantum feature map using custom kernel functions.
+
+import numpy as np
+from sklearn.metrics.pairwise import rbf_kernel
+from sklearn.svm import SVC
+
+# Define a custom kernel function to simulate a quantum feature map
+def custom_kernel(X, Y):
+    return rbf_kernel(X, Y, gamma=0.5)  # Using RBF kernel as an example
+
+# Train SVM with the custom kernel
+clf_qsvm = SVC(kernel=custom_kernel)
+clf_qsvm.fit(X_train, y_train)
+
+# Predict and evaluate
+y_pred_qsvm = clf_qsvm.predict(X_test)
+accuracy_qsvm = accuracy_score(y_test, y_pred_qsvm)
+print(f'Simulated QSVM Accuracy: {accuracy_qsvm * 100:.2f}%')
+

doc/src/week13/programs/qsvmqiskit.py

@@ -0,0 +1,39 @@
+from qiskit import BasicAer
+from qiskit.utils import QuantumInstance
+from qiskit.circuit.library import ZZFeatureMap
+from qiskit_machine_learning.algorithms import QSVC
+from sklearn.model_selection import train_test_split
+from sklearn.datasets import load_iris
+from sklearn.preprocessing import StandardScaler
+
+# Load and preprocess dataset
+dataset = load_iris()
+X = dataset.data
+y = dataset.target
+
+# For simplicity, we'll only classify between two classes
+X = X[y != 2]
+y = y[y != 2]
+
+# Standardize features
+scaler = StandardScaler()
+X = scaler.fit_transform(X)
+
+# Split into training and test sets
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+# Define quantum feature map
+feature_map = ZZFeatureMap(feature_dimension=4, reps=2)
+
+# Set up quantum instance
+quantum_instance = QuantumInstance(BasicAer.get_backend('qasm_simulator'), shots=1024)
+
+# Initialize QSVC
+qsvc = QSVC(quantum_instance=quantum_instance, feature_map=feature_map)
+
+# Train QSVC
+qsvc.fit(X_train, y_train)
+
+# Evaluate QSVC
+accuracy = qsvc.score(X_test, y_test)
+print(f'QSVC accuracy: {accuracy * 100:.2f}%')