如何在 Python 中为 CNN keras 绘制 ROC

问题描述

这是我的 CNN 通过库 Keras 的代码，我想绘制一个 ROC 并认为我需要在函数中执行它，就像我对精度图和损失图所做的那样。任何人都可以帮助我解决这个问题，因为我不确定如何去做并且目前正在挣扎。 roc 曲线的代码是否需要在函数中，如果不需要，我将如何对其进行编码

def CNN(imgs,img_labels,test_imgs,test_labels,stride):

    #Number of classes (2)
    num_classes = len(img_labels[0])

    #Size of image
    img_rows,img_cols=imgs.shape[1],imgs.shape[2]
   input_shape = (img_rows,img_cols,3)

    #Creating the model
    model = Sequential()

    #First convolution layer
    model.add(Conv2D(32,kernel_size=(3,3),activation='relu',input_shape=input_shape))

    #First maxpooling layer
    model.add(MaxPooling2D(pool_size=(2,2)))

    #Second convolution layer
    model.add(Conv2D(64,(3,activation='relu'))

    #Second maxpooling layer
    model.add(MaxPooling2D(pool_size=(2,2)))

    #Third convolution layer
    model.add(Conv2D(128,activation='relu'))

    #Third maxpooling layer
    model.add(MaxPooling2D(pool_size=(2,2)))

    #Convert the matrix to a fully connected layer
    model.add(Flatten())

    #Dense function to convert FCL to 128 values
    model.add(Dense(128,activation='relu'))

    #Final dense layer on which softmax function is performed
    model.add(Dense(num_classes,activation='softmax'))

    #Model parameters
    model.compile(loss='categorical_crossentropy',optimizer='adam',metrics=['accuracy'])

    #Evaluate the model on the test data before training your model
    score = model.evaluate(test_imgs,verbose=1)

    print('\nKeras CNN binary accuracy:',score[1],'\n')

    #The model details
    history = model.fit(imgs,shuffle = True,epochs=30,validation_data = (test_imgs,test_labels))

    #Evaluate the model on the test data after training your model
    score = model.evaluate(test_imgs,verbose=1)
    print('\nKeras CNN binary accuracy:','\n')

    #Predict the labels from test data
    y_pred = model.predict(test_imgs)
    Y_pred_classes = np.argmax(y_pred,axis=1) 
    Y_true = np.argmax(test_labels,axis=1)

    #Correct labels
    for i in range(len(Y_true)):
        if(Y_pred_classes[i] == Y_true[i]):
            print("The predicted class is : ",Y_pred_classes[i])
            print("The real class is : ",Y_true[i])
            break
        
    #The confusion matrix made from the real Y values and the predicted Y values
    confusion_mtx = [Y_true,Y_pred_classes]

    #Summary of the model
    model.summary()

    # summarize history for accuracy
    plt.plot(history.history['accuracy'])
    plt.plot(history.history['val_accuracy'])
    plt.title('model accuracy')
    plt.ylabel('accuracy')
    plt.xlabel('epoch')
    plt.legend(['Train','Validation'],loc='upper left')
    plt.show()
    # summarize history for loss
    plt.plot(history.history['loss'])
    plt.plot(history.history['val_loss'])
    plt.title('model loss')
    plt.ylabel('loss')
    plt.xlabel('epoch')
    plt.legend(['Train',loc='upper left')
    plt.show()

    return model,confusion_mtx

model,conf_mat = CNN(X_train,y_trainHot,X_test,y_testHot,1);

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）