留出驗(yàn)證集

x_val = x_train[:1000]
partial_x_train = x_train[1000:]

y_val = one_hot_train_labels[:1000]
partial_y_train = one_hot_train_labels[1000:]

訓(xùn)練模型

history = model.fit(partial_x_train,
                    partial_y_train,
                    epochs=20,
                    batch_size=512,
                    validation_data=(x_val, y_val))

繪制訓(xùn)練損失和驗(yàn)證損失

import matplotlib.pyplot as plt

loss = history.history['loss']
val_loss = history.history['val_loss']

epochs = range(1, len(loss)+ 1)

plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.legend()

plt.show()

訓(xùn)練損失和驗(yàn)證損失.png

繪制訓(xùn)練精度和驗(yàn)證精度

plt.clf() # 清空圖像

acc = history.history['acc']
val_acc = history.history['val_acc']

plt.plot(epochs, acc, 'ro', label='Training acc')
plt.plot(epochs, val_acc, 'r', label='Validation acc')
plt.title('Training and validation accuracy')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
plt.legend()

plt.show()

訓(xùn)練精度和驗(yàn)證精度.png

可以看出在第9輪之后開始過擬合。

重新開始訓(xùn)練一個模型

model.fit(x_train,
          one_hot_train_labels,
          epochs=9,
          batch_size=512
          )
results = model.evaluate(x_test, one_hot_test_labels)

print(results)

[0.9613658222680843, 0.7902938557966204]

字?jǐn)?shù)不夠，哈哈來湊！哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈