如何 plot 预测 model？

Question

I have output of my model y_pred and x_test . How can I easily see the results of it by plotting with plt?

Answer 1

here I am plotting results in 2 rows:

def visualiseResult(y_pred, x_test):
    fig, ax = plt.subplots(2, 5)
    for i in range(0, 10, 1):

        y_pred_base = y_pred[0][i]
        y_pred_aux  = y_pred[1][i]

        y_pred_base = y_pred_base.reshape(48,48)
        y_pred_aux  = y_pred_aux.reshape(48,48)

    #     visualizeImage(y_pred_base)
    #     visualizeImage(y_pred_aux)

        x1 = y_pred_base.reshape(48*48)
        y1 = y_pred_aux.reshape(48*48)

    #     print(list(x1).index(np.max(x1)))
    #     print(list(y1).index(np.max(y1)))

        x1 = list(x1).index(np.max(x1))
        x = round(x1/47)
        y = round(x1%48)

        row = int(i / 5)
        col = round(i%5)

        ax[row][col].imshow(x_test[i], cmap = 'gray')
        ax[row][col].scatter(x, y)
    fig.show()

Answer 2

Your question is vague, but I will try to help:

Here is an example for training history after fitting the model to see overfitting/underfitting

from matplotlib import pyplot as plt

date = datetime.now().strftime("%Y_%m_%d-%I:%M:%S_%p")  

history = model.fit(X_train, y_train,
                    epochs=200,
                    batch_size=16,
                    validation_data=(X_val, y_val),
                    callbacks = [checkpoint,earlystopping])
                    
# Visualize the training history to see whether you're overfitting.
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.title('Model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['TRAIN', 'VAL'], loc='lower right')
plt.savefig('saved_charts/training_history' + str(date) + '.png')
plt.show()

Since you are doing prediction/test, I am guessing you are trying to plot the accuracy results for that. I recommend a confusion matrix:

from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
from matplotlib import pyplot as plt
from matplotlib.collections import LineCollection
import numpy as np

def plot_confusion_matrix(cm, classes, normalize = False, title='Confusion matrix', cmap=plt.cm.Blues):
    #plots confusion matrix
    date = datetime.now().strftime("%Y_%m_%d-%I:%M:%S_%p")

    if normalize:
        cm = cm.astype('float') / cm.sum(axis =1)[:, np.newaxis]
        print("Normalized confusion matrix")
    else:
        print('Confusion matrix, without normalization')
    plt.imshow(cm, interpolation='nearest', cmap=cmap)
    plt.title(title)
    plt.colorbar()
    tick_marks = np.arange(len(classes))
    plt.xticks(tick_marks, classes, rotation=55)
    plt.yticks(tick_marks, classes)
    fmt = '.2f' if normalize else 'd'
    thresh = cm.max()/2. 
    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
        plt.text(j, i, format(cm[i,j], fmt), horizontalalignment="center", color="white" if cm[i,j] > thresh else "black")
    plt.ylabel('True Label')
    plt.xlabel('Predicted Label')
    plt.tight_layout()
    plt.savefig('saved_charts/confusion_matrix_' + str(date) + '.png')

y_pred = model.predict(X_test)
y_pred_label = [class_names[i] for i in np.argmax(y_pred, axis=1)]
y_true_label = [class_names[i] for i in np.argmax(y_test, axis=1)]
#plot the confusion matrix
cm = confusion_matrix(np.argmax(y_test, axis=1), np.argmax(y_pred, axis=1))
plot_confusion_matrix(cm, class_names, title = 'Confusion Matrix of Model')