pyqpanda_alg.QSVM.quantum_kernel_svm

Module Contents

Classes

QuantumKernel_vqnet

A help class to create Quantum Kernal Matrix.The evaluate function can be used in sklearn.svm .

class pyqpanda_alg.QSVM.quantum_kernel_svm.QuantumKernel_vqnet(batch_size: int = 100, n_qbits=None)

A help class to create Quantum Kernal Matrix.The evaluate function can be used in sklearn.svm .

Parameters

batch_size: int

Data number to build quantum kernel, default: 100.

n_qbits: int

number of quantum bits.

evaluate(x_vec: numpy.ndarray, y_vec: numpy.ndarray = None) → numpy.ndarray

Evaluation function to build quantum kernel.

Parameters

x_vec: ndarray

Train or test dataset features.

y_vec: ndarray

Train or test dataset labels.

Returns

out: ndarray

Kernal matrix.

Examples

import os
import numpy as np
import pyqpanda as pq
from sklearn.svm import SVC
import matplotlib
try:
    matplotlib.use('TkAgg')
except:
    pass
import matplotlib.pyplot as plt


from pyqpanda_alg.QSVM.quantum_kernel_svm import QuantumKernel_vqnet

from pyqpanda_alg import QSVM
data_path = QSVM.__path__[0]


def _read_vqc_qsvm_data(path):
    train_features = np.loadtxt(os.path.join(path, "dataset/qsvm_train_features.txt"))
    test_features = np.loadtxt(os.path.join(path, "dataset/qsvm_test_features.txt"))
    train_labels = np.loadtxt(os.path.join(path, "dataset/qsvm_train_labels.txt"))
    test_labels = np.loadtxt(os.path.join(path, "dataset/qsvm_test_labels.txt"))
    samples = np.loadtxt(os.path.join(path, "dataset/qsvm_samples.txt"))
    return train_features, test_features, train_labels, test_labels, samples
def qsvm_classification():
    train_features, test_features, train_labels, test_labels, samples = _read_vqc_qsvm_data(data_path)
    plt.figure(figsize=(5, 5))
    plt.ylim(0, 2 * np.pi)
    plt.xlim(0, 2 * np.pi)
    plt.imshow(
        np.asmatrix(samples).T,
        interpolation="nearest",
        origin="lower",
        cmap="RdBu",
        extent=[0, 2 * np.pi, 0, 2 * np.pi],
    )

    plt.scatter(
        train_features[np.where(train_labels[:] == 0), 0],
        train_features[np.where(train_labels[:] == 0), 1],
        marker="s",
        facecolors="w",
        edgecolors="b",
        label="A train",
    )
    plt.scatter(
        train_features[np.where(train_labels[:] == 1), 0],
        train_features[np.where(train_labels[:] == 1), 1],
        marker="o",
        facecolors="w",
        edgecolors="r",
        label="B train",
    )
    plt.scatter(
        test_features[np.where(test_labels[:] == 0), 0],
        test_features[np.where(test_labels[:] == 0), 1],
        marker="s",
        facecolors="b",
        edgecolors="w",
        label="A test",
    )
    plt.scatter(
        test_features[np.where(test_labels[:] == 1), 0],
        test_features[np.where(test_labels[:] == 1), 1],
        marker="o",
        facecolors="r",
        edgecolors="w",
        label="B test",
    )

    plt.legend(bbox_to_anchor=(1.05, 1), loc="upper left", borderaxespad=0.0)
    plt.title("samples dataset for classification")

    plt.show()
    samples_datasets_kernel = QuantumKernel_vqnet(n_qbits=2)
    samples_datasets_svc = SVC(kernel=samples_datasets_kernel.evaluate)
    samples_datasets_svc.fit(train_features, train_labels)
    samples_datasets_score = samples_datasets_svc.score(test_features, test_labels)

    print(f"Callable kernel classification test score: {samples_datasets_score}")

    samples_datasets_matrix_train = samples_datasets_kernel.evaluate(x_vec=train_features)
    samples_datasets_matrix_test = samples_datasets_kernel.evaluate(x_vec=test_features, y_vec=train_features)

    fig, axs = plt.subplots(1, 2, figsize=(10, 5))
    axs[0].imshow(
        np.asmatrix(samples_datasets_matrix_train), interpolation="nearest", origin="upper", cmap="Blues"
    )
    axs[0].set_title("samples training kernel matrix")
    axs[1].imshow(np.asmatrix(samples_datasets_matrix_test), interpolation="nearest", origin="upper", cmap="Reds")
    axs[1].set_title("samples testing kernel matrix")
    plt.show()


if __name__ == "__main__":
    qsvm_classification()