深度学习——Kears调用训练好的模型并进行评估绘制混淆矩阵

 实验代码：

import pickle as my_pickle
import matplotlib.pyplot as plt
from keras.utils import np_utils
from sklearn.metrics import confusion_matrix
import numpy as np
import keras
if __name__ == "__main__":
    # 加载数据集和标签
    T1 = my_pickle.load(open("data/dataset/1_train.p", "rb"))
    D1 = my_pickle.load(open("data/dataset/1_dev.p", "rb"))
    labels1 = my_pickle.load(open("data/dataset/4_lables_train.p", "rb"))
    labels2 = my_pickle.load(open("data/dataset/4_lables_dev.p", "rb"))

    Y_train = np_utils.to_categorical(labels1, 6)
    Y_test = np_utils.to_categorical(labels2, 6)

    # 加载训练好的模型
    model = keras.models.load_model('C1-100-0.66.hdf5')

    # 进行预测
    Y_pred = model.predict(D1)
    y_pred = np.argmax(Y_pred, axis=1)
    y_true = np.argmax(Y_test, axis=1)

    # 计算混淆矩阵
    cm = confusion_matrix(y_true, y_pred)
    classes = ['Class 0', 'Class 1', 'Class 2', 'Class 3', 'Class 4', 'Class 5']
    num_classes = len(classes)

    # 绘制混淆矩阵
    plt.figure(figsize=(8, 6))
    plt.imshow(cm, interpolation='nearest', cmap=plt.cm.Blues)
    plt.title('Confusion Matrix')
    plt.colorbar()

    tick_marks = np.arange(num_classes)
    plt.xticks(tick_marks, classes, rotation=45)
    plt.yticks(tick_marks, classes)

    # 在每个矩阵单元格中显示数值
    thresh = cm.max() / 2.
    for i in range(num_classes):
        for j in range(num_classes):
            plt.text(j, i, format(cm[i, j], 'd'),
                     horizontalalignment="center",
                     color="white" if cm[i, j] > thresh else "black")

    plt.tight_layout()
    plt.ylabel('True Label')
    plt.xlabel('Predicted Label')
    plt.savefig('confusion_matrix.png')

    plt.show()

运行结果：